Against Free-Form Formalism - NYU Law Review

Imaged with the Permission of N.Y.U. Law Review 

ARTICLES 

AGAINST FREE-FORM FORMALISM 

DAVID M. GOLOVE* 

Article II, Section 2 of the Constitution provides that the President shall have the 

power to make treaties by and with the consent of two-thirds of tie Senate. Yet, 

most of the international agreements into which the United States has entered over 

the last fifty years, including NAFTA and the WTO Agreement, have been concluded 

as congressional-executive agreements-a procedure by which the President 

submits an agreement to both houses of Congress for simple majority approvaL In 

an article in the Harvard Law Review entitled "Is NAFTA Constitutional," 

Professors Bruce Ackerman and David Golove attempted to provide this practice 

with constitutionalfoundations. Pointing to the decisive shift away from traditional 

isolationism and toward internationalism at the close of World War II, they argued 

the resulting "constitutional moment" transformed the meaning of the Treaty 

Clause and legitimized the congressional-executive agreement. Professors Ackerman 

and Golove also advanced a narrower argument for which it was unnecessary 

to rely on the full thrust of Professor Ackerman's theory of higher lawmaking. 

They claimed tha4 notwithstanding contrary practice during the nation's first 150 

years, the constitutional text is indeterminate and can plausibly be construed to support 

the congressional-executive agreement 

In a subsequent article; Professor Laurence Tribe sharply criticized this latter aspect 

of Professors Ackerman and Golove's artie, charging that their claim that die text 

is indeterminate reflects a dangerous "free-formn" approach to constitutional interpretation 

in which arguments are selectively chosen to read preordained conclusions 

without concern for fidelity to the text. In order to preserve interpretive 

objectivity and the ability of the Constitution to constrain government, Professor 

Tribe urged that constitutional interpretation, at least as to "architectral" provisions, 

should be based strictly on the original meaning of the text. He then sought 

to illustrate the proper application of his own "topological" model of textual interpretation 

by closely focusing on the Treaty Clause Developing an elaborate set of 

textual arguments, he claimed that there is only one plausible construction of the 

text: The Treaty Clause is exclusive and renders the congressional-executive agreement 

unconstitutional 

* Associate Professor of Law, Benjamin N. Cardozo School of Law, Yeshiva University 

(during the time much of this Article was written, of the University of Arizona College 

of Law). A.B., 1979, J.D., 1982, University of California, Berkeley; LLM., 1993, Yale University. 

Special thanks to Bruce Ackerman (who insists that I say "that he would gladly 

sign on to this Article as coauthor but for the fact that he has not done any of the work!"), 

Laurence Tribe, Toni Massaro, Susan Lewis, Mark Ascher, Christopher Eisgruber, Thomas 

Franck, Lewis Komhauser, Carol Rose, Scott Shapiro, Peter Spiro, Elliott Weiss, Rob Williams, 

participants in the Cardozo Law School faculty workshop I gave on the subject in 

Fall 1997, and to Michael Reisman, whose off-hand comment that he had found Professor 

Tribe's textual arguments persuasive rankled in the back of my mind until exorcised by the 

writing of this Article. 

1791


1792 

NEW YORK UNIVERSITY LAW REVIEW 

[Vol. 73:1791 

In this Article, Professor Golove responds to Professor Tribe on the latter's own 

terms by offering a serious textual and structural analysis of the Treaty Clause that 

supports its nonexclusivity. Professor Golove shows that the constitutional text is in 

fact indeterminate and that, contrary to Professor Tribe's claims, textualism cannot 

render a singularly persuasive construction of the Treaty Clause. By analyzing 

each of Professor Tribe's arguments, Professor Golove shows that equally strong 

formal arguments can be constructed in favor of the nonexclusive reading. Professor 

Golove thus seeks to demonstrate by illustration that textualism is just as open 

to manipulation as the interpretive methodologies that Professor Tribe decries and, 

given the pervasive ambiguities in the text, is generally incapable of yielding unique, 

objective resolutions to constitutional disputes, even those over concrete provisions 

of the text. Only by systematically ignoring these equally plausible formalist 

counterarguments was Professor Tribe able to reach his favored reading of the 

Treaty Clause. In the final analysis, Professor Tribe's article reflects free-formism 

in its most paradoxical form: free-form formalism. 

[Professor] Tribe is a skilled litigator, and he does not lightly set 

aside arguments in favor of his positions. 1 

INTRODUCTION 

Professor Tribe is indeed a skilled litigator, but far more than 

that, he is one of the great advocates in our constitutional history. No 

one would seriously deny the grace and power of his rhetoric and oratory. 

Nor, in my opinion, is there any other constitutional scholar who 

has made a larger or more admirable contribution to the advancement 

of our collective project. Like Justice Brennan, he is the builder of an 

inspired edifice of doctrine that, in Professor Dworkin's sense, makes 

our Constitution "the best it can be ' 2 and lays the foundation for its 

further realization far into the future. 

Yet, my admiration for his prodigious accomplishments notwithstanding, 

there are times when even the most accomplished among us 

go astray. And in those moments, elegant prose and formidable rhetorical 

skills, called into action by the imperatives of internecine battle, 

can make a deadly combination. Incisive critique, persuasive 

arguments, and arresting insights fly by so fast that we are swept along 

without a chance to scrutinize their deeper merits. But when the dust 

settles-if it ever settles before most of us just move on-there is an 

accounting that ought to be done, a searching assessment of what has 

been said, the critiques, the arguments, and the insights, for they may 

1 Ronald Dworkin, The Arduous Virtue of Fidelity: Originalism, Scalia, 'Tribe, and 

Nerve, 65 Fordham L. Rev. 1249, 1261 (1997). 

2 Ronald Dworkin, Law's Empire 62 (1986) [hereinafter Dworkin, Law's Empire]; see 

also Ronald Dworkin, Freedom's Law 11 (1996) [hereinafter Dworkin, Freedom's Law] 

(declaring that when more than one construction "fits" the relevant constitutional materials, 

interpreters are to "decide on their own which conception does most credit to the 

nation").


December 1998] 


have been less persuasive than at first appeared and they may have 

wider implications that reveal deeper flaws in our methodologies and 

commitments. 

With these preliminary remarks, I will turn to my own assessment 

of Professor Tribe's most recent contribution to the Harvard Law Review, 

3 for I fear it suffers from some of the flaws I have identified and 

provides an occasion to examine critically some of the interpretive 

methodologies he now favors. Not that I am wholly unimplicated in 

the dispute; indeed, I am a participant, one of the objects of Professor 

Tribe's heated rhetoric. For that reason, if not on general principles 

alone, I invite others to assess my own arguments, to find where I 

have missed the obvious or perhaps unconsciously overlooked the 

subtle in my effort to defend my own role. Yet, my thesis is that, despite 

having turned a microscope to the Constitution and scoured the 

text for any supporting evidence, Professor Tribe has not made a single 

argument beyond the familiar that supports his exclusive reading 

of the Constitution's Treaty Clause, nor given us a single reason for 

accepting his exclusivist view as the only plausible construction of the 

text. By demonstration, however, he has given us powerful reasons 

for skepticism about formalist interpretive methodologies that require 

a blinkered, single-minded focus on the original meaning of the text. 

The main target of Professor Tribe's ire is what he calls "freeform 

method" in constitutional interpretation. In his view, freeformism 

has, regrettably, become all too common among constitutional 

scholars and is particularly marked in the work of Professor 

Bruce Ackerman of the Yale Law School. Although unusually clever, 

Professor Ackerman is among the most egregious purveyors of the 

practice, and his much-discussed theory of higher lawmaking and informal 

constitutional amendments very nearly reeks of it.4 Exhibit 

Number One for the prosecution is a lengthy article Professor 

Ackerman and I coauthored in the Harvard Law Review, entitled "Is 

NAFTA Constitutional" s We sought there to explain the constitu- 

3 See Laurence H. Tribe, Taking Text and Structure Seriously. Reflections on Free- 

Form Method in Constitutional Interpretation, 108 Harv. L Rev. 1221 (1995) [hereinafter 

Tribe, Taking Text]. 

4 Professor Ackerman's theory is given its fullest presentation in 1 Bruce Ackerman, 

We the People: Foundations (1991), and in 2 Bruce Ackerman, We the People: Transformations 

(1998). 

5 Bruce Ackerman & David Golove, Is NAFrA Constitutional, 108 Harv. L Rev. 

799 (1995) (tracing rise of congressional-executive agreement and so-called interchangeability 

doctrine, whereby Congress may approve by majority vote any international agreement 

that President could submit for two-thirds approval by Senate under Article II, 

Section 2 of Constitution); see also Bruce Ackerman & David Golove, Is NAFTA Constitutional 

(1995) (reprinting same as book).



[Vol. 73:1791 

tional foundations for a long-settled practice whereby the United 

States enters into international agreements on the authority of an act 

of Congress rather than by the constitutionally specified procedure 

requiring the advice and consent of two-thirds of the Senate. 6 

Although the vast bulk of our post-War international commitments 

have been concluded in this form, two recent examples are particularly 

dramatic: the North American Free Trade Agreement 

(NAFTA) 7 and the Agreement Establishing the World Trade Organization 

(WTO Agreement). 8 

Professor Tribe fiercely disagrees with our constitutional views. 

Unembarrassed by the fact that they are supported nearly unanimously 

by contemporary scholars of foreign relations law 9 and have 

been reflected in actual constitutional practice on thousands of occasions 

during the past fifty-plus years' 0 -indeed, even though it requires 

a volte face from his own earlier constitutional views"- 

Professor Tribe believes that our constitutional claims can only be 

supported by free-form interpretive methods. Nevermind that our express 

purpose was to show that the constitutionality of the so-called 

congressional-executive agreement procedure rests upon an extratextual 

act of popular sovereignty in the wake of the unprecedented devastation 

of World War II and the widespread belief that the Treaty 

Clause had been at least partly responsible for the tragic events following 

the rejection of the Treaty of Versailles. Professor Tribe nevertheless 

believes that free-formism is conspicuously on display 

throughout our argument. Applying the precept that the most effec- 

6 See U.S. Const. art. II, § 2, cI. 2. 

7 See North American Free Trade Agreement Implementation Act, Pub. L. No. 103- 

182, 107 Stat. 2057 (1993) (authorizing by two-house majority vote the North American 

Free Trade Agreement, Dec. 17, 1992, 32 I.L.M. 605 (1993)). 

8 See Uruguay Round Agreements Act, Pub. L. No. 103-465, 108 Stat. 4809 (1994) 

(authorizing by two-house majority vote the Agreement Establishing the World Tltade Organization, 

opened for signature Apr. 15, 1994, 33 I.L.M. 1144 (1994)). 

9 See, e.g., Restatement (Third) of Foreign Relations Law § 303(2) (1987) [hereinafter 

Restatement] (upholding constitutionality of interchangeability doctrine); id. § 303 cmt. e 

(elaborating on same); Louis Henkin, Foreign Affairs and the United States Constitution 

215-18 (2d ed. 1996) (same); Memorandum of Law from Professors Bruce Ackerman, 

Abram Chayes, Kenneth Dam, Thomas Franck, Charles Fried, David Golove, Louis 

Henkin, Robert Hudec, John H. Jackson, Harold Hongju Koh & Myres McDougal to 

Members of Congress and Executive Branch Officials (Nov. 11, 1994) (on file with the New 

York University Law Review) (same); see also Ackerman & Golove, supra note 5, at 805 

n.12 (citing authorities). 

10 For a number of examples, see the discussion in Ackerman & Golove, supra note 5, 

at 875-907; see also Executive Agreements, 14 Whiteman Digest § 22, at 196, 210 (noting 

that vast bulk of our international commitments since World War II have been concluded 

as congressional-executive agreements). 

n1 See Laurence H. Tribe, American Constitutional Law § 4-5, at 228 n.18 (2d ed. 1988) 

[hereinafter Tribe, Constitutional Law] (endorsing interchangeability doctrine).




tive teaching is by example, he therefore sets out to demonstrate how 

a rigorous and detailed textual exegesis of the Treaty Clause should be 

done. When it is over, it establishes, conclusively in his view, that Professor 

Ackerman and I, and all the leading scholars in the field-not 

to mention the many presidents, senators, and representatives from 

both parties who have consistently upheld the interchangeability doctrine 

over many yearslz-are wrong: The Treaty Clause is subject to 

only one plausible construction. It is wholly exclusive; there is no such 

constitutional animal as the congressional-executive agreement. 

Perhaps even more startling than this conclusion is Professor 

Tribe's understanding of what a rigorous, as opposed to a free-form, 

interpretive methodology requires. I will not be the only one surprised 

by his turn towards strict textualism in constitutional interpretation. 

1 3 Distinguishing between what he calls "architectural" and 

"aspirational" provisions of the Constitution, he insists that the former, 

which include the Treaty Clause, must be read as if frozen in time 

circa 1787 while the latter may (though they need not necessarily) be 

given more room to evolve and expand.' 4 In any case, at least as to 

his vaguely specified architectural provisions, Professor Tribe selfconsciously 

joins ranks with Justice Scalia ' 5 and takes up the cudgels 

for a formalist interpretive methodology that relies solely upon text 

and insists that the interpreter's goal is simply to uncover its original 

meaning. Indeed, like Justice Scalia, he believes that the intent of the 

Framers is largely irrelevant, except as it is instantiated in the text, and 

he is highly skeptical about the use of precedent and practice to inform 

constitutional interpretation. 16 

12 For numerous examples, see Ackerman & Golove, supra note 5, at 861-907. 

13 Even Professor Tribe has excused those who might find "some amusement" in his 

endorsement of textualism. Tribe, Taking Text, supra note 3, at 1248 n.91 (citing Sanford 

Levinson, Authorizing Constitutional Text: On the Purported Twenty-Seventh Amendment, 

11 Cost. Commentary 101, 105 (1994)); see also Jordan Steiker et al., Taking Text 

and Structure Really Seriously- Constitutional Interpretation and the Crisis of Presidential 

Eligibility, 74 Tex. L. Rev. 237, 243-52 (1995) (parodying, I believe, Professor Tribe's original 

meaning textualism). 

14 Tribe, Taking Text, supra note 3, at 1247 & n.90; see also infra notes 35-37 and accompanying 

text (elaborating on Tribe's distinction between architectural and aspirational 

provisions). 

15 Indeed, the two have recently canvassed the commonalities, and differences, in their 

perspectives. See Antonin Scalia, Common-Law Courts in a Civil-Law System: The Role 

of United States Federal Courts in nterpreting the Constitution and Laws, in A Matter of 

Interpretation: Federal Courts and the Law 3 (Amy Gutmann ed., 1997) [hereinafter 

Scalia, Common-Law Courts]; Laurence H. Tribe, Comment, in A Matter of Interpretation, 

supra, at 65 [hereinafter Tribe, Comment]; Antonin Scalia, Response, in A Matter of 

Interpretation, supra, at 129, 133-43 [hereinafter Scalia, Response]. 

16 See infra note 38 and accompanying text.



[Vol. 73:1791 

Rather than addressing the meta-interpretive questions raised by 

Professor Tribe's views directly, I focus on the details of his elaborate 

textual arguments. Through close analysis of those arguments, I shall 

attempt both to liberate the Treaty Clause from his narrow exclusive 

construction of its language and to demonstrate the rigid, unconvincing 

character of his formalistic interpretive methodology. In the process, 

moreover, I hope to illustrate some of the central defects in 

Professor Tribe's and Justice Scalia's unrelenting focus on text and 

original meanings to the exclusion of more nuanced, inclusive interpretive 

methodologies. 

In the first two parts of this Article, I draw the contours of the 

present dispute between Professor Tribe on the one hand and Professor 

Ackerman and me on the other, and then rehearse what I take to 

be the main textual arguments on both sides of the question. In my 

view, those arguments are ultimately indeterminate, leaving the original 

meaning proponent ample room to read the Treaty Clause as 

either exclusive or nonexclusive. In Part III, I then undertake a painstaking 

assessment of each and every argument that Professor Tribe 

urges in favor of his insistent claim that the exclusive view is the only 

plausible interpretation of the text. 17 What emerges is that beyond the 

rather straightforward arguments that I contend make the exclusive 

reading one plausible construction of the text, Professor Tribe has 

failed to make a single persuasive argument to support his view. And 

while he makes a number of thought-provoking general observations 

about the nature of textual interpretation, he fails to heed his own 

advice, disregarding the dictates of his own "topological" model of 

interpretation that charges us with taking text, structure, and architecture 

seriously. Clever though many of the long series of arguments he 

17 I make a determined effort to address all of Professor Tribe's arguments addressed 

to the interpretation of the Treaty Clause. I do not address all of his arguments against 

Professor Ackerman's larger constitutional views nor the other interpretive disputes between 

Professors Tribe and Ackerman, and between Professors Tribe and Amar. Compare 

Tribe, Taking Text, supra note 3, at 1243-47, 1288-92 (criticizing Ackerman's and Amar's 

interpretive methods generally, and their interpretations of Article V specifically), with 

Bruce Ackerman, Higher Lawmaking, in Responding to Imperfection: The Theory and 

Practice of Constitutional Amendment 63, 71-82 (Sanford Levinson ed., 1995) (describing 

non-Article V process of "higher lawmaking" whereby "We the People," acting as national 

sovereign, may amend Constitution), Akhil Reed Amar, The Consent of the Governed: 

Constitutional Amendment Outside Article V, 94 Colum. L. Rev. 457 (1994) (defending 

reading of Constitution that allows amendment outside Article V), and Akhil Reed Amar, 

Philadelphia Revisited: Amending the Constitution Outside Article V, 55 U. Chi. L. Rev. 

1043 (1988) (same). Although I touch on some of the main points, especially as they relate 

to my larger argument, see infra notes 403; 415-33, and accompanying text, I leave a fuller 

defense to those who have made the arguments under siege, who are fully capable of defending 

their own views. For Professor Ackerman's most extended textual defense of his 

construction of Article V, see 2 Ackerman, supra note 4, at 71-92.




puts forth may be, ironies abound, and, in the end, the arguments collapse 

in contradiction and indeterminacy, leaving us with no choice 

but to condemn him for the very sin which he sought to lay on the 

heads of Professor Ackerman and me: free-form method in constitutional 

interpretation. Only in his case, this method takes on a most 

paradoxical form: free-form formalism. 

Still, the important point is not really about the rigor of Professor 

Tribe's arguments. As our preeminent scholar of constitutional doctrine, 

he is at least as well placed as anyone to offer a convincing illustration 

of how a textualist methodology can yield a singularly 

persuasive construction of constitutional text. All the more so, then, 

when he offers an extended textual exegesis of a relatively concrete 

provision that was uniformly construed in accord with his view during 

the first 150 years of the Republic. If his arguments prove unpersuasive 

and ultimately self-defeating even in this context, this surely says 

something of wider significance about the weaknesses of the textualist 

project as a whole: The pervasive textual ambiguities in the Constitution, 

exemplified by the Treaty Clause, will most often prove unresolvable 

by application of purely formalist interpretive techniques. 

Once having made good on these admittedly bold claims, I turn 

to some larger theoretical concerns. Professor Tribe's central thesis 

appears to be that Professor Ackerman's theory of constitutional moments 

is itself a deplorable example of free-formism, one that needs to 

be extirpated in order to save the field from loose interpretive techniques 

and shoddy scholarship. But this is a category error of the first 

order. Whatever else it may be, Professor Ackerman's theory is not 

an interpretive theory at all, although it certainly has interpretive implications. 

Rather, it is a theory of informal higher lawmaking, of constitutional 

transformation outside the confines of Article V. 

Ironically, and quite contrary to Professor Tribe's charged rhetoric, it 

is motivated in significant measure precisely by the need to liberate 

ourselves from reliance on free-formism to justify our basic political 

institutions. 

Given Professor Tribe's preeminent stature in the community of 

constitutional scholars, judges, and advocates, I believe this extended 

response to his article warranted. In the final analysis, his recent arguments 

disserve the Treaty Clause; disserve the wider field of foreign 

relations law; disserve the project of defending the Constitution and 

realizing its full potential as the charter of our political community, a 

project for which he has long been at the forefront; and disserve even 

those committed to textualism and original meaning methodologies in 

constitutional law.



[Vol. 73:1791 

I 

ANATOMY OF T= 

DisPuTE 

Let us begin with what is in dispute and with what has already 

been said, both of which can be summarized in a few pages. At stake 

is how the United States enters into binding international commitments. 

There are basically three procedures which are used, although 

only one, the treaty procedure, is specified in the constitutional text. 18 

Under the treaty mode, the President negotiates an agreement and 

then seeks the advice and consent of the Senate. In order to obtain 

constitutional authority to make the treaty internationally binding, he 

must convince two-thirds of the Senators present to give their approval. 

Under the congressional-executive agreement procedure, in 

contrast, the President may conclude an agreement under the authority 

of an act of Congress, which, like all other legislation, must be approved 

by simple majorities in both houses. Finally, the President 

sometimes concludes unilateral executive agreements on his sole 

authority. 

The present dispute concerns only the second category, the socalled 

congressional-executive agreement. 1 9 This category, in turn, 

breaks into two sub-types: those agreements Congress authorizes the 

President ex ante to conclude in accordance with whatever guidelines 

it chooses to enact and those Congress, mimicking the Senate in its 

treaty mode, authorizes ex post after the President has negotiated an 

agreement and sought legislative approval. 20 NAFTA and the WTO 

18 See U.S. Const. art. II, § 2, cl. 2. 

19 As a sidelight, our accounts have importantly different implications for the constitutional 

underpinnings of the third category, the unilateral executive agreement, as well, but 

that is a secondary, though not necessarily less important, concern in the present controversy. 

For further discussion, see infra notes 358, 376; 327-29, 340-43, and accompanying 

text. 

20 There is a fourth type of agreement, sometimes known as the executive agreement 

ancillary to a treaty. In these cases, the President, without further authority, concludes a 

special agreement that is more or less explicitly called for in a treaty that has already 

received the Senate's advice and consent. In effect, this is a kind of ex ante agreement, 

approved by the Senate at the time it gives its advice and consent to the treaty. There is, 

no doubt, a cognate executive agreement ancillary to a congressional-executive agreement. 

By a parity of reasoning, moreover, although there is little if any practice to confirm the 

view, it may be the case that the Senate could give its ex ante consent authorizing the 

President to conclude treaties on particular subjects in accordance with its stipulated requirements. 

To my knowledge, it has never done so, and in the famous dispute over the 

Hague arbitration treaties around the turn of the century, the Senate adamantly refused to 

go even half way there. For further discussion, see John Bassett Moore, Treaties and Executive 

Agreements, 20 Pol. Sci. Q. 385 (1905). Some senators did worry that the 1943 

Connally Resolution, which called for the creation of an international organization to keep 

the peace, might be construed as the Senate's ex ante consent to the then forthcoming 

United Nations Charter. See 89 Cong. Rec. 9066 (1943) (remarks of Sen. Connally) (noting 

concerns of senators to this effect). In any case, the constitutional question raised, as in



AGAINST FREE-FORM FORMIALISM 

Agreement are important, but by no means unique, examples of the 

latter. The longstanding majority view, and the settled practice, is that 

treaties and congressional-executive agreements, whether ex ante or 

ex post, are wholly interchangeable. 21 This means that the President 

has the option for any agreement to seek the Senate's consent by twothirds 

vote or the approval of Congress by simple majority vote.2 z Of 

course, there is nothing to prevent the Senate from insisting as the 

price of its cooperation that the President opt for the treaty route. 

But, likewise, the House may insist on the congressional-executive 

agreement form by refusing to enact legislation necessary to implement 

the agreement as a treaty. 

Since "[c]onstitutional doctrine to justify Congressional- 

Executive agreements is not clear or agreed," 23 Professor Ackerman 

and I sought to provide the needed constitutional foundations. It nevertheless 

remains an open question whether our efforts have furthered 

or hindered that goal. For the main burden of our extended argument 

was to show that the best and most widely accepted account 24 rests 

upon false historical premises and that the real origin of the practice is 

to be found in the decisive internationalist movement led by President 

Roosevelt during World War IM25 In short, we painstakingly sought to 

demonstrate that the conventional historical accounts developed in 

the 1940s were wrong in claiming that the congressional-executive 

agreement had roots going back to the earliest days after the founding 

and had always been a consistent part of the nation's agreementmaking 

practices. Contrary to the claims of leading scholars and executive 

branch officials in the 1940s, we argued that during the first 150 

the case of ex ante congressional-executive agreements, would be the scope of the Senate's 

power to delegate its treaty responsibilities to the President. Cf. Field v. Clark, 143 U.S. 

649, 693-94 (1892) (upholding congressional delegation of authority). 

21 See Ackerman & Golove, supra note 5, at 805 n.12 (citing authorities); supra notes 9- 

10 (same). 

22 This common formulation of the interchangeability doctrine is not entirely precise. 

In fact, the doctrine asserts that Congress may approve any international agreement submitted 

by the President that falls within one of its enumerated (or implied) powers in 

Article I, Section 8 of the Constitution or elsewhere. For further discussion and some 

additional refinements, see infra notes 28-49, 56, and accompanying text. 

23 Henkin, supra note 9, at 216. 

24 Indeed, it is the only account that has seriously been offered in the literature. 

25 We did note that important, though circumscribed, inroads were made upon the 

traditional understanding during the early New Deal period, when limited uses of the ex 

ante congressional-executive agreement were first made. See Ackerman & Golove, supra 

note 5, at 845-61. Further, we left open the degree to which this more limited practice 

might have been legitimized as part of the sweeping domestic transformations that were 

realized in 1937 of the commerce and other powers. We nevertheless did make unmistakably 

clear our view that whatever status these changes may have achieved, they in no way 

approached the modem interchangeability doctrine. See id. at 851-56, 860-61.


1800 NEW YORK UNIVERSITY LAW REVIEW [Vol. 73:1791 

years of our history, there was no practice which supported interchangeability 

or even the existence of the congressional-executive 

agreement form in its modern guise. 26 The Framers' intent, moreover, 

was consistent only with an exclusive reading. 27 Having thus cast 

doubt on the existing foundations for the practice, we offered a more 

controversial alternative account: The people in the mid-1940s, exercising 

their popular sovereignty through an act of informal higher lawmaking, 

had transformed the meaning of the Treaty Clause to 

constitute the modem congressional-executive agreement and render 

it interchangeable with the traditional treaty. 28 

Although Professor Tribe mostly agrees with our history, he most 

assuredly rejects our theory. This might well have led one to expect 

him to focus on his theoretical objections to Professor Ackerman's 

larger project, and he does make some arguments, which I come to 

later, 29 against the idea of informal higher lawmaking. But by far the 

main subject of his remarks is interpretive. He vehemently objects to 

our claim that the text of the Treaty Clause is indeterminate and need 

not necessarily be read to require that all international agreements be 

26 See id. at 813-45. 

27 See id. at 808-13. 

28 See id. at 861-96. There are actually two versions of the interchangeability doctrine, 

one stronger and the other weaker. We explicitly defended the weaker version, but the 

stronger is probably also consistent with our argument. The weaker claims only that Congress 

has the power under the Necessary and Proper Clause to approve international 

agreements that are within the reach of its enumerated and implied powers. The stronger 

claims, without support in the text, that Congress has the power to approve any international 

agreement. For all practical purposes, there seems to be no difference between the 

two. Congress's foreign affairs powers are very broad, perhaps plenary. See Henkin, supra 

note 9, at 64-80 (detailing source and breadth of Congress's foreign affairs power); Louis 

Henkin, The Treaty Makers and the Law Makers: The Law of the Land and Foreign Relations, 

107 U. Pa. L. Rev. 903, 928-29 (1959) (arguing that any agreement that is beyond 

Congress's enumerated powers would nevertheless fall within its implied foreign affairs 

powers). Thus, the two versions tend to collapse into one another. If some gap still remains, 

moreover, it might be filled by the President's independent powers. Some have 

argued that because the President and Congress together are the repositories of American 

sovereignty, all agreements they make together must be within their combined powers. 

See Henkin, supra note 9, at 216. But this argument overlooks the fact that there is another 

relevant alternative to be considered-the President and the Senate acting as the 

treaty makers. It is theoretically possible, then, that an agreement would be beyond both 

Congress's, the President's, and their combined authority. While the textual arguments I 

make in this Article are consistent only with the weaker version, it is by no means clear 

that the constitutional compromise reached in 1945 is not best read as supporting the 

stronger view. In that case, it would be completely clear that the interpretive points Professor 

Tribe raises are irrelevant to our argument. There is, however, no need to consider 

this possibility here. 

29 See infra Part IV.




validated by two-thirds of the Senate. 30 This claim was only a minor 

theme in our overall argument and was offered in the hope that it 

would make it easier for some to accept the implications of Professor 

Ackerman's theory of informal higher lawmaking. Although the 

movement of 1945 was, in our view, sufficient to create an informal 

amendment of the Constitution, such a sweeping proposition was unnecessary 

to establish the constitutionality of the congressionalexecutive 

agreement. Because the text is indeterminate, our argument 

could rest upon a weaker claim: A constitutional movement of 

the kind we described justifies the adoption of a construction of the 

text in conflict with long-settled constitutional practice and original 

intent, so long as that construction can claim a plausible basis in the 

text. Although the textual argument was quite clearly unnecessary to 

our main point, Professor Tribe believes it reflects a free-form interpretive 

methodology that infects our entire argument, as well as Professor 

Ackerman's wider theory, and consequently warrants an 

extended protest 1 He thus undertakes to develop his own elaborate 

interpretive model for constitutional law and to illustrate its proper 

and rigorous application through detailed examination of the Treaty 

Clause. 

Professor Tribe prefaces his interpretive foray by developing an 

intriguing, extended metaphor between the fields of topology and 

constitutional law. 3 2 Topologists investigate the nature of certain continuously 

transforming objects in space, seeking to describe what alterations 

of their shapes leave their fundamental properties intact and 

which transform them into essentially new objects. Bending and 

stretching may not fundamentally alter the structure of a geometric 

configuration, but tearing or cutting would. For Professor Tribe, the 

governmental system established by the Constitution is in some ways 

like a multidimensional geometric object and, like it, the integrity of 

its architecture can be fundamentally altered by interpretive moves 

that amount to tearing or cutting, not just bending and stretching. 

"[T]he consequences of any such tearing may be that we end up with a 

30 See Tribe, Taking Text, supra note 3, at 1278-79 (concluding that "textual and structural 

considerations leave no genuine doubt as to the exclusivity of the Treaty Clause," and 

charging that we strained to find ambiguity in text). 

31 According to Professor Tribe, our 

NAFTA article... presents a significant threat to the whole enterprise of constitutional 

dialogue and decisionmaking-a threat implicit in [Professor 

Ackerman's] earlier works but made manifest here. The danger arises from a 

facile treatment of constitutional text and structure and a free-form approach 

to saying what they mean .... 

Id. at 1233. 

32 See id. at 1235-49.



[Vol. 73:1791 

system different in very basic ways from that envisioned by the Constitution. 

'33 That is precisely the consequence, he believes, of accepting 

the constitutional validity of the congressional-executive 

agreement. 

For Professor Tribe, this conclusion arises not from the combined 

effect of the intentions of those who framed and ratified the Constitution, 

of the text, and of the understandings during the first 150 years of 

our history. Instead, he insists that his conclusion can be-and indeed 

only should be-drawn on the basis of the "original meaning" of the 

text, which, he thinks, can bear only one plausible construction 4 

Three basic commitments underlie this turn towards originalism in its 

textualist garb. First, he argues that the Constitution's provisions 

come in two basic types, "architectural" and "aspirational. '' 35 The former, 

which include the structural provisions of the Constitution, 

"ought to be given as fixed and determinate a reading as possibleone 

whose meaning is essentially frozen in time insofar as the shape, 

or topology, of the institutions created is concerned. '36 In contrast, 

aspirational provisions might best be read through evolving lenses. 3 7 

Second, the unenacted intentions of a lawmaker, whether a legislature 

or the framers of a constitutional text, are largely irrelevant to the 

interpretive task. What matters is only what they said, not what they 

hoped or expected or assumed the consequences of their lawmaking 

would be. 38 Finally, precedent has only a very limited role to play in 

33 Id. at 1237. 

34 Id. at 1242 n.66. 

35 Id. at 1247 n.90. 

36 Id. at 1247. 

37 See id. Professor Tribe, however, has not been entirely consistent in his attitude 

towards the text. For example, despite his frequent affirmations of the text's primacy and 

of the imperative to seek "the best reading .. identified in terms of interpretive canons 

that are as immune as we can make them from the pushes and pulls of our own policy 

predilections," id. at 1279, he suggests at other points that the text exerts only a much 

looser constraint on the interpreter. It is only that "nothing irreconcilable with the text can 

properly be considered part of the Constitution." Tribe, Comment, supra note 15, at 77; 

see also Tribe, Taking Text, supra note 3, at 1237 (describing his topological model which, 

taken literally, allows for bending and stretching, just not cutting or tearing text). Of 

course, there is a world of difference between these views. If his searching efforts at Treaty 

Clause exegesis suggest a commitment to the former, then the glibness with which he affirms 

modem Commerce Clause jurisprudence, the shift in substantive due process doctrine 

from Lochner v. New York, 198 U.S. 45 (1905), to the contemporary privacy 

decisions, and the expansion of the President's unilateral agreement-making powers (as 

well as his past interpretive efforts in a variety of other areas), see infra notes 420-23, 449- 

50, and accompanying text, suggest the latter. Despite these apparent inconsistencies, I 

shall take him at his word and assume that he endorses the stricter view. 

38 See Tribe, Taking Text, supra note 3, at 1242 n.66 (declaring that concerning "consultation 

of the Framers, it should be axiomatic that it is enacted law-whether in the form of 

a statute or a constitution-that governs, never the unenacted intentions of any lawgiver");


December 1998] AGAINST FREE-FORM FORMALISM 

1803 

constitutional interpretation. The fact that the governmental departments 

have repeatedly and for lengthy periods engaged in conduct 

that cannot be squared with the best interpretation of the text simply 

represents usurpation, not grist for the evolutionist's interpretive 

mill. 3 9 But giving an abstract account of his interpretive model is only 

Professor Tribe's first task. The main part of his argument is a prodigious 

effort to "illustrate[ ]" his "approach to constitutional text, strucsee 

also Tribe, Comment, supra note 15, at 65-66 (stressing primacy of text's meaning over 

lawmakers' expectations or intentions). Only to the limited extent that the latter would 

point us toward "the linguistic frame of reference within which the people to whom those 

words or phrases were addressed would have 'translated' and thus understood them" might 

they be relevant in determining the original meaning of the text the lawmakers actually 

adopted. Id. at 65. 

39 See Tribe, Taking Text, supra note 3, at 1281-82. In addition to these three commitments, 

Professor Tribe's "topological" model enjoins us to beware the hazards of construing 

the sounds of constitutional silences, see id. at 1241, to avoid the pitfall of mistaking 

gaps in the Constitution's text as holes in constitutional space to be filled in at will, see id. 

at 1239-45, and to attend to how the entities created by the Constitution connect and interlock, 

see id. at 1248-49. It enjoins us as well to take not only text but structure and architecture 

seriously, "the pattern and interplay in the governmental edifice that the 

Constitution describes and creates, and in the institutions and practices it propels." Id. at 

1236. 

As with regard to the text, Professor Tribe is not wholly consistent in his attitude 

towards the use of history. In the past, he has been unwilling to articulate his own methodological 

commitments, expressing skepticism about the possibility of defending a global 

account, including textualism. See Laurence H. Tribe, Constitutional Choices 3-6 (1985) 

[hereinafter Tribe, Choices] (finding "all legitimating theories not simply amusing in their 

pretensions but, in the end, as dangerous as they are unconvincing"). Try as I may, however, 

I am unable to discern a coherent view other than textualism that could underlie his 

Treaty Clause argument. Most tellingly, he eschews the use of historical evidence in determining 

the meaning of the Treaty Clause, even though that evidence would decisively bolster 

his case. That is one of the central points of the article by Professor Ackerman and 

me, and with our history, if nothing else, Professor Tribe fully concurs. See Tribe, Taking 

Text, supra note 3, at 1230-31, 1270, 1280-81 (acknowledging that post-War rise of 

congressional-executive agreement marked break with prior, longstanding constitutional 

understanding). On the other hand, he at one point affirms that "constitutional interpretation 

would certainly be robbed of much if it were conducted in an historical vacuum-or 

even through historical lenses that could see only up to the point of a constitutional provision's 

adoption and not a moment beyond." Id. at 1280; see also id. at 1280-81 (affirming 

importance of early practice extending back to nation's founding). At another, he offhandedly 

makes passing reference to the understandings during the first century-and-a-half to 

support one of his textual claims. See id. at 1270; see also infra notes 223-24 and accompanying 

text (discussing page 1270 at length). And at still others, he supports the President's 

textually doubtful sole organ power and his power to conclude unilateral executive agreements 

by reference to early practice. See id. at 1255, 1265; see also infra notes 303-18 and 

accompanying text (discussing source and scope of sole organ power). Notwithstanding 

these apparent inconsistencies, I have assumed that Professor Tribe's basic position, at 

least as to the Treaty Clause, is textualist. Any other explanation would be inconsistent 

with the main thrust of his argument and would seem to trivialize the interpretive dispute 

between us.



[Vol. 73:1791 

ture, and history" 40 by an extended analysis of the Treaty Clause. 

Here, his three interpretive commitments-to the unchanging quality 

of architectural provisions, to the irrelevance of original intent, and to 

the limited value of historical precedent-lead him to embrace an interpretation 

of the Treaty Clause that obliterates the accepted constitutional 

basis for most of the international agreements concluded by 

this nation over the past half-century. It is somewhat embarrassing, 

given the zeal with which he now holds that view, that his own interpretive 

efforts have shifted quite dramatically on several past occasions, 

ranging from endorsing interchangeability, to various middle 

ground positions, to his current strictly exclusivist view. 41 Indeed, despite 

the sharpness of our disagreement over the WTO Agreement, 

Professor Tribe's position during the debate, and only a few months 

before his article was published, was actually remarkably close to our 

own. He publicly advised the President and the Senate that only the 

most monumentally important agreements, such as the WTO Agreement, 

require the sanction of two-thirds of the latter body. All other 

agreements, even important but less foundational bargains, could be 

approved by Congress. 42 In our article, we pointed out the instability 

of that reading from both the perspectives of history and of text and 

argued that the only plausible textual constructions were the ex- 

40 Tribe, Taking Text, supra note 3, at 1276. 

41 See Tribe, Constitutional Law, supra note 11, § 4-5, at 228 n.18 (endorsing interchangeability); 

Letter from Laurence H. Tribe, Professor, Harvard Law School, to Sen. 

Robert Byrd (July 19, 1994) (on file with the New York University Law Review) [hereinafter 

Tribe Letter to Sen. Byrd of July 19, 1994] (suggesting that only extremely important 

agreements need be submitted as treaties to Senate); Memorandum from Laurence H. 

Tribe, Professor, Harvard Law School, to Walter Dellinger, Assistant Attorney General, 

Office of Legal Counsel, et al. 8 (Oct. 5, 1994) (on file with the New York University Law 

Review) [hereinafter Tribe Memo of Oct. 5, 1994] (same); GATT Implementing Legislation: 

Hearings on S. 2467 Before the Senate Comm. on Commerce, Science, and Tansportation, 

103d Cong. 301 (1994) (prepared statement of Laurence H. Tribe, Professor, 

Harvard Law School) [hereinafter GATT Hearings] (widening net to include NAFTA and 

other important congressional-executive agreements); Tribe, Taking Text, supra note 3 (arguing 

for exclusivist view under which congressional-executive agreement form is unconstitutional). 

There is, of course, nothing wrong with changing one's view. Indeed, Professor 

Ackerman and I made some claims during the debate on the WTO Agreement that I 

would not now defend. See Ackerman & Golove, supra note 5, at 812 & n.45 (providing 

different interpretation of early comment by Madison than we had during debate on WTO 

Agreement, as Professor Tribe has pointed out, see Tribe, Taking Text, supra note 3, 1264 

n.146). A radical shifting of views such as Professor Tribe's, however, does at least raise 

doubts about the clarity of the text in question. 

42 See Tribe Memo of Oct. 5, 1994, supra note 41, at 8 (noting that if "one looks at the 

agreements other than NAFrA that have a remotely comparable impact [to that of the 

WTO Agreement], one will only find the United Nations Treaty [sic] itself and perhaps the 

establishment of the North Atlantic Treaty Organization," and implying constitutionality 

of all other congressional-executive agreements); see also GATT Hearings, supra note 41, 

at 301 (similar).



1805 

clusivist view or the modem interchangeability doctrine. 4 3 In response, 

Professor Tribe abandoned his earlier view, as it were, biting 

the bullet and announcing with yet increased fervor his new-found attachment 

to the largely archaic exclusivist camp. 4 

43 See Ackerman & Golove, supra note 5, at 922: 

Both the modem and traditional readings have a clarity and elegance to 

them-either (as modems think) Congress can create binding international obligations 

whenever it thinks it 'necessary and proper' under Article 1, or (as was 

generally believed before Versailles) Article I is never a source of this power 

and the Senate must alvays give its advice and consent. 

44 The consequence of his current view is that the congressional-executive agreement 

form, not any particular congressional-executive agreement, is unconstitutional, and while 

he obscures the point, see Tribe, Taking Text, supra note 3, at 1234-35 & 1234 n.47, 1265 

(attempting to imply that only ex post agreements are affected), this applies to both cx ante 

and ex post agreements. This result is breathtaking in its potential implications: It means 

not only that NAFTA, the WTO Agreement, and a host of other foundational ex post 

agreements are presumptively unconstitutional, but that the vast bulk of all agreements the 

United States has entered during the past fifty years are as well, since approximately 90% 

of our post-War commitments have been concluded as ex ante congressional-executive 

agreements. See Executive Agreements, 14 Whiteman Digest § 22, at 210; supra note 10 

and accompanying text. 

Given the enormity of this view, Professor Tribe unsurprisingly attempts to soften the 

blow, but the points he musters in mitigation are neither convincing nor particularly comforting. 

He notes first that some congressional-executive agreements actually received 

more than two-thirds support in the Senate, and, while carefully avoiding endorsing the 

argument, he suggests that this might cure any constitutional defect. See Tribe, Taking 

Text, supra note 3, at 1227 & n.18,1276. The reason for his equivocation is evident: The ex 

ante choice of the applicable voting rule may well influence the ultimate outcome, and it 

cannot be assumed that an agreement that generated a two-thirds majority as a 

congressional-executive agreement would necessarily have won the same support had it 

been submitted as a treaty. Moreover, Powell v. McCormack, 395 U.S. 486 (1969), at least 

arguably stands for the proposition that a congressional vote premised on the incorrect 

assumption that a simple majority rule applies must be invalidated even though there were 

enough affirmative votes to prevail under the properly applicable supermajority rule. See 

id. at 508 (refusing to validate vote in excess of two-thirds to exclude Representative 

Powell from House because vote -was taken under misapprehension that simple majority 

vote was sufficient when in fact two-thirds rule for expulsions applied). More importantly, 

Professor Tribe intimates that many of the agreements concluded under congressional authorization 

may well have been within the President's unilateral authority in any case, and 

hence valid on that basis. See Tribe, Taking Text, supra note 3, at 1269, 1276-77. This 

portentous expansion of contemporary and historical understandings of the scope of the 

President's unilateral powers is hardly comforting. It is precisely the dangers posed by 

such broad conceptions of independent executive authority that have been the preoccupation 

of foreign relations scholars for decades. In any case, Professor Tribe does not explain 

how he squares this view with his commitment to the frozen-in-time character of architectural 

provisions. It is scarcely imaginable that he thinks such broad unilateral powers accord 

with the best interpretation of the text of 1787. For further discussion, see infra notes 

423; 374-77 and accompanying text. None of this, moreover, addresses the potentially devastating 

impact that requiring senatorial advice and consent for large numbers of international 

agreements may have on the conduct of our foreign relations in the future.



[Vol. 73:1791 

II 

ANATOMY OF THE TREATY CLAUSE 

Before examining Professor Tribe's wide-ranging textual claims, 

it will be useful to provide an overview of the nature of the interpretive 

dispute between us. This can be accomplished in two ways. First, 

in the mathematical spirit introduced by Professor Tribe, I begin with 

a "Venn" diagram. This will help situate the dispute in the larger textual 

context of Articles I and II. Second, I will set out in summary 

form what I take to be the basic textual arguments on both sides of 

the debate. This will serve a number of functions. First, it will demonstrate 

the heavy burden Professor Tribe has assumed in insisting on 

the exclusive reading as the sole plausible construction of the text. 

Second, it will set the stage for my claim that Professor Tribe has not 

provided an ounce of additional support for his preferred construction 

beyond some straightforward and familiar arguments I articulate here. 

Finally, it will provide a road map that will help us to avoid getting 

lost in the maze even as we consider the dizzying array of constitutional 

provisions that the dispute over the Treaty Clause requires us to 

consult. 

Despite our disagreements, Professor Tribe is right in one respect: 

Until now, defenders of the congressional-executive agreement have 

not paid much attention to the text. Spurred by Professor Tribe's determined 

textualist challenge, I begin by filling this gap. 

A. The Treaty Clause in Textual Perspective 

The Treaty Clause, in its elusive simplicity, provides that the 

President "shall have Power, by and with the Advice and Consent of 

the Senate, to make Treaties, provided two thirds of the Senators 

present concur. ' 45 In Illustration 1, the circle on the right represents 

the entire set of international agreements that the United States may 

conclude, or to put it somewhat differently, the federal agreementmaking 

power; the circle on the left represents the entire set of congressional 

powers over foreign affairs, which are largely set out in Article 

I. At issue is how these two sets of powers are interrelated. We 

can easily identify three possible constructions-two extreme views 

and one middle position. The first claims that the Article II treaty 

power is comprehensive and is exclusively vested in the President and 

the Senate. As a consequence, the treaty power is coextensive with 

Areas B and C, and Congress's Article I powers do not extend even to 

agreements the subject matter of which falls within Area B. This is 

45 U.S. Const. art. II, § 2, cl. 2.




Professor Tribe's view. 46 The second view takes the contrary approach. 

The treaty power does not extend to any subject matter falling 

within the zone of Congress's Article I powers marked by Area B. 

Thus, the President and the Senate can conclude only those agreements 

falling within Area C-those matters that are in any case beyond 

Congress's reach. Area B, in contrast, belongs exclusively to 

Congress. As we shall see, this view was articulated early on by 

Thomas Jefferson and memorialized in his influential Manual of Parliamentary 

Practice. 47 

In contrast to these two polar positions, the third view claims the 

middle ground. Like both views, it acknowledges the Senate's exclusive 

power over Area C. And like the first view, it acknowledges the 

Senate's power over agreements falling in Area B. But the Senate's 

power over Area B is not exclusive; acting under its Article I powers, 

Congress may, as in the second view, approve agreements falling in 

Area B as well. In adopting this middle position, I defend the modem 

consensus: The Senate has plenary power over all international agreements, 

but exclusive authority only over those falling outside the 

scope of Congress's legislative authority. The interchangeability thesis 

is simply this: Because Congress's powers over foreign affairs are 

46 Actually, Professor Tribe's position is slightly different than I have indicated in the 

text. He believes that some agreements are not significant enough to rise to the level of an 

Article 11 "treaty." See infra Part IH.D.2. Although he denies the implication, this would 

appear to mean that the treaty power is not fully coextensive with the agreement-making 

power, and so some portion of Areas B and C would not be covered by the Senate's powvers. 

The gap belongs neither to the Congress nor to the Senate but to the President acting 

alone. See infra Part mII.D.2. As will become evident, this claim is wholly fallacious. See 

infra Part llI.D.2. 

47 See Thomas Jefferson, A Manual of Parliamentary Practice § 52 (1801), reprinted in 

Constitution, Jefferson's Manual and Rules of the House of Representatives, H.R. Doc. 

No. 104-272, at 115, 298-99 (1997). According to Jefferson, the Framers must have meant 

"to except those subjects of legislation in which it gave a participation to the House of 

Representatives." Id. § 52, at 299. Admittedly, though, it is uncertain whether Jefferson 

would have allowed Congress the power to approve agreements, only added the House as 

an additional participant along with two-thirds of the Senate, or denied altogether that the 

agreement-making power extends to subjects within Congress's legislative authority and 

required that those subjects be regulated purely through domestic statutes. See id. I assume 

the first alternative, but if I am wrong, the only point affected is the attribution to 

Jefferson. In any case, the somewhat awkward latter two positions, with different variations, 

were held by a number of early authorities. See, e.g., 1 Charles Henry Butler, The 

Treaty-Making Power of the United States §§ 300-316 (1902) (discussing early debates); 

Treaties, 5 Moore Digest § 735, at 164 (reprinting response by Secretary of State Calhoun 

to assertion of this kind in Senate Foreign Relations Committee report that recommended 

rejecting a tariff reciprocity treaty on ground that it infringed on Congress's power over 

foreign commerce). For Hamilton's vigorous attack on these positions during the Jay 

Treaty controversy, see Alexander Hamilton, The Defence Nos. 36-38 (Jan. 2, 6, 9, 1796), 

reprinted in 20 The Papers of Alexander Hamilton 3, 5-10, 13-34 (Harold C. Syrett ed., 

1974) (writing as Camillus).



ILLUSTRATION 1 

Congress's Foreign A (1 The Federal Agreement 

Affairs Powers I llill i!Making Power 

(virtually) plenary, 48 the two circles are entirely (or nearly entirely) 

overlapping. Thus, in virtually all cases, either the Senate or the Congress 

can approve an international agreement; the two procedures are 

for all practical purposes interchangeable. 

By pairing Jefferson's exclusivist view along with Professor 

Tribe's, I do not mean to suggest that it is a plausible third contender. 

As I explain below, however, my reasons for rejecting it are historical, 

not textual. Analytically, Jefferson's polar reading presents Professor 

Tribe's view with a serious challenge-to which, as we shall see, he 

never responds. 

B. The Basic Textual Arguments 

I turn now to a summary of the argument in favor of the nonexclusive 

or middle view. It has two parts: first, the affirmative case for 

the congressional-executive agreement, and second, the negative case 

against the exclusivity of the Treaty Clause, which includes responses 

to what I take to be the two principal arguments in favor of Professor 

Tribe's view. 

The affirmative argument rests principally on Congress's Article 

I, Section 8 enumerated powers and the Necessary and Proper 

Clause. 49 In Article I, the Framers provided amply for Congress's authority 

over matters touching on foreign affairs, giving it, most importantly, 

the power to regulate foreign commerce, 50 to declare war, 51 

and to raise and support armies. 52 There is every reason to suppose 

that Congress might find it necessary in exercising one of these powers, 

for instance the power over foreign commerce, to authorize the 

President to conclude an agreement with a foreign nation, say one 

48 For further discussion, see supra note 28; infra notes 49-52, 298, 301-05, and accompanying 

text. 

49 See U.S. Const. art. I, § 8, cl. 18. 

50 See id. art. I, § 8, cl. 3. 

51 See id. art. I, § 8, cl. 11. 

52 See id. art. I. § 8. cl. 12.




that lowers tariffs. In that event, given McCulloch v. Maryland's S3 expansive 

reading of the Necessary and Proper Clause, Congress 

would seem to have the authority to pass, as "necessary and proper 

for carrying into Execution" 55 the foreign commerce power, a law authorizing 

the President, ex ante or ex post, to conclude the agreement. 

The approval of an international agreement is simply a means, like 

chartering a bank or making greenbacks legal tender, 5 6 for achieving 

the substantive aims which the enumerated powers allow. Hence, the 

interchangeability doctrine: Congress may approve any agreement 

the subject matter of which falls within its substantive powers enumerated 

or implied in Article I, Section 8 (or elsewhere in the text). Additional 

support for this reading of Congress's necessary and proper 

powers can be found in the Supremacy Clause.5 The Supremacy 

Clause instructs that treaties are to be the "Law of the Land," Ss suggesting 

the legislative character of the procedure through which they 

are promulgated. 5 9 

Of course, the fact that the Necessary and Proper Clause is sufficiently 

roomy to contain the congressional-executive agreement does 

not mean that the modem procedure is necessarily constitutional. 

Perhaps, in Justice Holmes's elegant phrase, it is still ruled out "by 

some invisible radiation from the general terms" 60 of the Treaty 

Clause. Hence, the negative argument against the exclusivity of the 

Treaty Clause, which responds to the two principal arguments-one 

textual and the other structural-in favor of the exclusive view. 

Consider first the textual argument for exclusivity, which rests on 

the maxim expressio unius est exclusio alterius. 61 According to this 

view, we should draw a negative inference as to Congress's power to 

approve agreements from the fact that the Treaty Clause is the only 

provision in the Constitution that explicitly specifies a procedure 

53 17 U.s. (4 Wheat.) 316 (1819). 

54 See id. at 421 ("Let the end be legitimate, let it be within the scope of the constitution, 

and all means which are appropriate, which are plainly adapted to that end, which are 

not prohibited, but consist with the letter and spirit of the constitution, are 

constitutional."). 

55 U.S. Const. art. I, § 8, cl. 18. 

56 See infra notes 112, 238, and accompanying text. 

57 See U.S. Const. art. VI, cl. 2. 

58 Id. 

59 For further discussion of the Necessary and Proper Clause, see infra Part III.B.1. For 

further discussion of the Supremacy Clause, see infra note 250. 

60 Missouri v. Holland, 252 U.S. 416, 434 (1920) (Holmes, J.) (rejecting any implicit 

limits on treaty power emanating from Tenth Amendment). 

61 Professor Tribe relies upon the maxim heavily, although he also relies on a number 

of other textual and structural arguments. I consider all of these arguments in Part I1, 

infra.


1810 


[Vol, 73:1791 

through which the federal government may make international agreements. 

62 The Framers assigned the federal treaty power to the 

President and the Senate; the expression of one thing is the exclusion 

of others. 

In response, the textual argument for the nonexclusive view begins 

with the Compact Clause, which prohibits states from entering 

into treaties, alliances, or confederations, but permits them, with Congress's 

consent, to conclude agreements or compacts with foreign 

states. 63 This clause demonstrates that the Framers had no aversion to 

congressional involvement in agreement-making: It expressly gives 

Congress, not the Senate or the President, responsibility for supervising 

agreements and compacts negotiated by states. This suggests that 

the Framers created the advice and consent procedure to give the 

President a special option, not to rule out resort to the usual congressional 

procedures. 64 Furthermore, early practice, thrice affirmed by 

the Supreme Court, 65 supports the President's authority to make unilateral 

executive agreements incident to his foreign affairs powers. 

But it stands to reason that if notwithstanding the Treaty Clause the 

President may make (some) agreements without the advice and consent 

of the Senate, then Congress has a parallel power to approve 

agreements incident to its enumerated powers over foreign affairs. 

Thus, a consistent exclusivist view would deny not only Congress's 

power to approve agreements but the President's power to shortcircuit 

the Treaty Clause as well. Yet, as we shall see, even Professor 

Tribe is unwilling to go that far, repeatedly affirming the President's 

power to act independently of the Senate. 6 Nor, finally, should we be 

daunted by the text's contrasting explicitness in Article II and silence 

in Article I. The Framers had good reason to be explicit about the 

Senate's role because senatorial advice and consent is an extraordinary 

procedure that could not otherwise be inferred from the constitutional 

text. This stands in sharp contrast to congressional 

authorizations which could be inferred from the implied powers doctrine 

written explicitly into the text in the Necessary and Proper 

Clause. 67 

62 The only other procedure for making agreements specified in the text applies strictly 

to the states. See U.S. Const. art. I, § 10, cls. 1, 3 [collectively referred to hereinafter as the 

"Compact Clause"]. 

63 See id. 

64 For further discussion, see infra notes 378-79 and accompanying text. 

65 See infra notes 329, 351-55, and accompanying text. 

66 For further discussion, see infra Part III.D.2. 

67 For further discussion, see infra text accompanying note 100.




Second, consider the structural argument in favor of the exclusive 

reading. It claims support in what it takes to be the Framers' purposes 

for assigning the advice and consent power to the Senate. On this 

view, it was no accident that the Framers assigned the advice and consent 

function to the Senate, in which the states, large and small, are 

equally represented. This tended to equalize their influence over the 

content of treaties. The exclusion of the House probably reflected the 

concern that its short term of office and large number of members 

made it unsuitable for the task. Most importantly, by including a twothirds 

supermajority voting requirement, the Framers effectively protected 

regional and sectional interests that might otherwise have been 

prejudiced by a treaty supported by a simple majority. In light of 

these apparent purposes, it would hardly be a matter of indifference 

whether treaties are to be approved by two-thirds of the Senate or by 

majorities in both houses. 68 

For the nonexclusivist, text and structure yield a different explanation 

for the Treaty Clause. To be sure, the Framers must have included 

the Senate in part because of the equal representation of the 

states, just as they had included the Senate in normal lawmaking functions 

in part for this reason. And they excluded the House because 

the short terms of Representatives and their large numbers might 

make their participation problematic in some cases, given the imperatives 

of secrecy, dispatch, and long-term perspective in international 

negotiations. However, the two-thirds requirement was added not to 

protect sectional or minority interests, but to make up for the occasional 

unavoidable absence of the House. As Madison put it, "a concurrence 

of two-thirds at least is made necessary, as a substitute or 

compensation for the other branch of the legislature, which, on certain 

occasions, could not be conveniently a party to the transaction." 69 In 

this respect, the Treaty Clause follows an established pattern in the 

text for dealing with the absence of one of the normal lawmaking organs: 

To substitute for the absence of the President in the case of a 

6 For further discussion, see infra notes 273-75 and accompanying text. It might further 

be supposed that the two-thirds rule reflected the Framers' wish to make entering 

treaties difficult. The supermajority requirement would encourage the young country to 

avoid entangling alliances. Although this is certainly a plausible interpretation, I do not 

discuss it explicitly in the text hereafter. The same arguments that respond to the federalism 

argument-that the two-thirds rule embodies special protections for the states-also 

reply to the isolationism argument. 

69 James Madison, Letter of Helvidius No. 1 (Aug.-Sept. 1793), in 6 The Writings of 

James Madison 138, 148 (Gaillard Hunt ed., 1906) [hereinafter Helvidius No. 1]. I quote 

Madison's views not as evidence of the Framers' intent, but of a leading Framers construction 

of the text itself. Indeed, I assume that Madison held a contrary "intent," in the sense 

that, as a participant in the founding and state ratifying conventions, he was fully aware 

that one of the central purposes of the two-thirds rule was to protect sectional interests.



veto, the text requires a two-thirds vote in both houses; 70 to make up 

for his or her nonparticipation in constitutional amendment proposals, 

71 two-thirds of both houses must give their assent; 72 to fill in for 

the House in treatymaking, two-thirds of the Senate, as well as the 

President, must consent. 73 The House's participation in agreementmaking, 

however, is not inappropriate in all cases. It was enough to 

give the President the option of consulting the Senate alone, when 

secrecy, dispatch, or other considerations so required. When those 

concerns were absent, there was no reason to preclude him from seeking 

the approval of the Congress as a whole and no reason to read the 

Treaty Clause as commanding the contrary. 74 We have especially 

powerful reasons, moreover, to give a narrow construction to a provision 

derogating from the usual bicameral procedure designed to empower 

broad-based majorities and protect our liberties. 75 

By briefly rehearsing these arguments, I do not mean to suggest 

that there is no more to be said on either side. My claim is simply that 

for all his efforts, Professor Tribe has failed to contribute anything 

new or compelling to the textual case for exclusivity and that from the 

purely textualist point of view he espouses, neither his reading nor my 

own-nor Jefferson's for that matter-can oust the others from the 

field of reasonable construction. As an initial matter, none can claim 

the mantle, "best reading of the text." But even for a strict adherent 

of original meaning textualism, constitutional interpretation is never 

ab initio, never conducted entirely in a historical vacuum. And when 

the interpreter is faced with a long-settled practice, deeply woven into 

the fabric of day-to-day governmental practices and widely perceived 

as essential to the conduct of government, and when that practice has 

been accepted or acquiesced in by presidents, senates, and houses of 

representatives for over half a century, the situation is most assuredly 

not ab initio. The procedure will be entitled to a strong presumption 

in favor of its constitutionality, and even the textualist will have no 

choice but to uphold the practice if there is a plausible and persuasive 

70 See U.S. Const. art. I, § 7, cls. 2, 3. 

71 See Hollingsworth v. Virginia, 3 U.S. (3 DalI.) 378, 378-379, 382 (1798) (affirming 

that President has no role in amendment proposal process). 

72 See U.S. Const. art. V. 

73 See I.N.S. v. Chadha, 462 U.S. 919, 955 & n.21 (1983) (citing U.S. Const. art. II, § 2, 

cl. 2). Likewise, to substitute for the absence of the House and the President, conviction 

on impeachment, the sole prerogative of the Senate, requires the concurrence of two-thirds 

of the members present. See id. (citing U.S. Const. art. I, § 3, cl. 6). For further discussion, 

see infra note 186. 

74 See Henkin, supra note 9, at 494 n.156 (suggesting similar reading of Framers' purposes). 

The Compact Clause provides further support for this view, as discussed infra note 

378. 

75 For further discussion, see infra notes 183-88 and accompanying text.




construction of the text that permits it.76 We should therefore expect 

that justices who, like Professor Tribe, place heavy emphasis on text 

will have little trouble upholding the congressional-executive agreement, 

should a justiciable controversy raising the question ever reach 

the Court. 77 If trouble is to be expected, it will more likely come from 

other quarters-from justices slavishly devoted to original intent and 

to subsequent history reflecting original understandings, for these justices 

will no doubt find themselves feeling ill-at-ease when they discover 

the gaping chasm between modem practice and the Framers' 

designs. Should they strike down this basic strut of our constitutional 

practices, or bend their principles to the will of the people circa 

194578 If the constitutional history developed by Professor 

Ackerman and myself is right, they are likely to uphold the practice, 

uncomfortable though they may feel, perhaps only partially aware of 

the larger historical patterns into which their actions fit. 9 

76 Despite Professor Tribe's heavy reliance on it, the Court's decision in Chadha is not 

to the contrary. The Court repeatedly made clear that the legislative veto was entitled to a 

presumption of validity. See Chadla, 462 U.S. at 944, 951-52. The Court simply believed 

that the original intent and the text were not open to more than one plausible and persuasive 

construction, characterizing them as "crystal clear," id. at 958-59, as providing an -unmistakable 

expression," id. at 959, and as "[e]xplicit and unambiguous," id. at 945. In 

addition, presidents had from the beginning protested against the legislative veto on constitutional 

grounds. See id. at 942 n.13 (noting that 11 presidents from Wilson to Reagan 

objected to legislative veto). In contrast, the Senate has long acquiesced in the 

congressional-executive agreement form-indeed, a congressional-executive agreement 

cannot be approved without the consent of at least a majority of the Senate. See 

Ackerman & Golove, supra note 5, at 889-916 (providing history). 

77 Perhaps Justice Scalia fits this description, but despite some rather strong statements 

denigrating the use of history in favor of text, see Scalia, Common-Law Courts, supra note 

15, at 29-37, and affirming textualism, see id. at 23-25, in practice he seems far more willing 

than Professor Tribe to consider at least the early precedents. His recent opinion in Printz 

v. United States, 117 S. Ct. 2365 (1997), among many others, is a case in point. See id. at 

2370-71 (considering early federal statutes that imposed certain administrative and judicial 

duties on state courts in determining whether federal government may require state law 

enforcement officers to perform background checks on purchasers of handguns). In response 

to Professor Tribe's comment on his recent Tanner Lectures, Justice Scalia placed 

some distance between himself and Professor Tribe on a closely related point, emphasizing 

the unduly narrow character of Tribe's version of original meaning methodology. See 

Scalia, Response, supra note 15, at 133-34 (highlighting differences with Professor Tribe on 

role of contemporary understandings of text); see also Scalia, Common-Law Courts, supra 

note 15, at 38 (explaining his view on role of contemporary understandings). 

78 The Court may yet get a chance to consider the issue. Suit was recently brought 

challenging the constitutionality of NAFTA. See Made in the U.S.A. Found. v. United 

States, No. CV-98-PT-1794-M (N.D. Ala. filed July 13, 1998). 

79 In this respect, however, Chadha presents a contrary indicator. There, the Court 

resisted the force of the New Deal revolution in refusing to uphold the legislative veto. 

See Chadha, 462 U.S. at 959.



III 

ANATOMY OF AN ORIGINAL MEANING TEXTUALIST 

ARGUMENT: THE HAZARDS OF 

CLEVERNESS UNBOUND 

I turn now to Professor Tribe's extended efforts to buttress the 

exclusive reading by a host of novel arguments. By examining them in 

depth, I hope to aid in the liberation of the Treaty Clause from the 

clutches of his exclusivist textual construction. I also hope to demonstrate 

the flaws in an interpretive methodology that, despite its own 

pretensions, ultimately rests heavily on formalistic mechanical rules 

and word games rather than on serious contextual and structural 

analysis. 

In considering his many substantive arguments, 80 I have for ease 

of exposition divided them into two somewhat arbitrary categories: 

arguments resting principally on purely linguistic or textual concerns 

and arguments drawing their force from larger structural considerations. 

I have divided these categories again into affirmative arguments 

for the exclusive reading and negative arguments seeking to 

undermine the case for the nonexclusive view. I have already briefly 

described what I believe are the two principal arguments for the exclusive 

reading-the arguments from expressio unius and from the 

states' rights orientation of the Framers (the structural federalism argument). 

Surprisingly, Professor Tribe says little about these arguments 

beyond conclusory assertion. This may implicitly reflect his 

recognition that they are insufficient on their own to ground a definitive 

original meaning construction-hence, his concentration on other 

arguments which he hopes, taken together, will do the trick. 

80 At my request, Professor Tribe was gracious enough to provide me with detailed 

comments on the manuscript of this Article after it was accepted for publication. See Letter 

from Laurence H. Tribe, Professor, Harvard Law School, to David M. Golove, Professor, 

University of Arizona College of Law (Apr. 30, 1998) (on file with the New York 

University Law Review) [hereinafter Tribe Letter to Golove of Apr. 30, 1998]. In his letter, 

he pointed out that at a few points I had attributed to him arguments which he had not 

intended to make. See id. In response, I have in most cases accepted his comments at face 

value and modified my arguments accordingly. In one instance, however, I comment in a 

footnote on why I believe his text strongly lends itself to my interpretation, but I still avoid 

in my text any attribution of that view to Professor Tribe. See infra note 104. In another 

instance, I have continued to attribute to Professor Tribe a view which he has now disowned 

because, notwithstanding his subjective intentions, I believe that the argument is 

unequivocally articulated in his original text. I do indicate in a footnote that Professor 

Tribe has now stated that in fact he did not intend to make that argument, see infra note 

105, and I have added some additional material responding to how he now articulates his 

position on that point, see infra notes 124-38 and accompanying text.




A. Affirmative Textual Arguments 

In this section, I address Professor Tribe's claims grounded in the 

expressio unius maxim and then turn to his arguments based on the 

Appointments 81 and Compact Clauses.8 

1. Expressio Unius and its Limits 

As I have already made clear, the principal textual argument in 

favor of the exclusive reading is, in my view, the canon expressio unius 

est exclusio alterius, and Professor Tribe repeatedly invokes it as a definitive 

point in favor of his construction of the text. But the logical 

force of this maxim is limited, and the strength of any inference of 

exclusivity it offers depends significantly upon context. Even when 

the linguistic context is particularly supportive, its application is often 

tempered by other considerations.83 When the wording of the text 

renders the inference less compelling, as in the Treaty Clause, the persuasive 

force of the maxim is at a minimum. 

It has long been recognized that drawing the expressio unius inference 

can be a risky venture. 84 In constitutional adjudication, if the 

Court has sometimes applied the canon, it has often explicitly rejected 

it.85 Commentators, too, have singled it out for criticism as resting 

upon unrealistic assumptions and have pointed out the mistaken consequences 

it may produce. 86 Even classical formulations of the principle 

have been quick to warn against its unthinking application, 

81 U.S. Const. art. H, § 2, cl. 2. 

82 Id. art. I, § 10, cls. 1, 3. 

83 For example, consider the Court's unwillingness in Springer v. Philippine Islands, 277 

U.S. 189,206 (1928), to apply the maxim in the face of the most compelling textual considerations. 

The inference must yield, said the Court, "whenever a contrary intention on the 

part of the law-maker is apparent." Id. (construing Organic Act for Philippine Islands 

during period of United States colonial rule); see also infra notes 175-77 and accompanying 

text (discussing Springer). 

84 Professor Tribe himself recognizes that expressio unius arguments "have their limits." 

Tribe, Taking Text, supra note 3, at 1273. 

85 See, e.g., Springer, 277 U.S. at 206 (rejecting application of erpressio unius); The 

Legal Tender Cases, 79 U.S. (12 Wall.) 457, 544-47 (1870) (same); Anderson v. Dunn, 19 

U.S. (6 Wheat.) 204, 232-33 (1821) (same). 

86 See, e.g., William N. Eskridge, Jr., Dynamic Statutory Interpretation 278-79 (1994) 

(observing that legislators do not always, and sometimes cannot, draft in accordance with 

interpretive canons such as expressio unius); Richard A. Posner, The Federal Courts: Crisis 

and Reform 282 (1985) (criticizing expressio unius canon as relying on mistaken assumption 

that all legislative omissions are deliberate and warning that misuse of canon 

may defeat legislative objectives); Cass R. Sunstein, Interpreting Statutes in the Regulatory 

State, 103 Harv. L. Rev. 405, 455-56 (1989) (warning that "[t]he expressio unius canon 

should not be used mechanically").



cautioning that "it leads to safe and satisfactory conclusions" only 

when the appropriate conditions are present. 87 

Professor Tribe has not exercised the necessary caution in arguing 

that the maxim entails his reading of the Treaty Clause. Begin by returning 

to the Venn diagram introduced in Illustration 1. The 

Jeffersonian position, it will be recalled, reads Congress's Article I 

powers over foreign affairs as exclusive of any overlapping Article II 

power in the President and the Senate to make treaties on matters 

subject to congressional regulation. This view confronts Professor 

Tribe's confident expressio unius argument with an immediate embarrassment. 

In all critical respects, the Jeffersonian view is just the mirror 

image of Professor Tribe's and can rely as readily upon the 

expressio unius inference. For it is not at all clear ab initio which 

power should be taken as "expressed" and which as "excluded." Consider, 

for example, Congress's foreign commerce powers and the Senate's 

treaty powers. Professor Tribe would have us deem the treaty 

power "expressio" and then exclude any congressional power to regulate 

commerce by approving agreements. But why is Jefferson's opposite 

reading any less persuasive Congress is expressly given the 

power to regulate foreign commerce in Article I; hence, the treaty 

power should not extend to any agreement that encroaches upon its 

authority over that subject. Expressio unius est exclusio alterius. 88 To 

be sure, Jefferson's view has been overwhelmed by the march of 

events; 89 but this is true of Professor Tribe's reading as well. Given 

Jefferson's stature-and the bluntness of expressio unius as an inter- 

87 J.G. Sutherland, Statutes and Statutory Construction 410 (Chicago, Calaghan & Co. 

1891). 

88 Nor for Jefferson was it any objection that this might leave precious little for the 

President and the Senate. See Jefferson, supra note 47, § 52, at 299 ("The less the better, 

[some] say .... The Constitution thought it wise to restrain the Executive and Senate from 

entangling and embroiling our affairs with those of Europe."). On the other hand, he 

seems to have believed that this was not actually so. See id. (pointing out that most matters 

would still properly remain as subjects of treaties). This latter point reflects a constricted 

view of the scope of Congress's legislative powers that can no longer be sustained 

today. 

89 See Treaties, 5 Moore Digest § 735, at 164 (reprinting note by Secretary of State 

Calhoun stating that "[i]f this be the true view of the treaty-making power, it may be truly 

said that its exercise has been one continual series of habitual and uninterrupted infringements 

of the Constitution"). Nonetheless, arguments of this kind were still current as late 

as the early part of this century and formed the principal basis of Senator Lodge's constitutional 

objections to the Covenant of the League of Nations. See the report of the Senate 

Foreign Relations Committee, of which Senator Lodge was the Chairman, S. Rep. No. 66- 

176, pt. 1, at 5-6 (1919) (claiming that Article X of Covenant unconstitutionally purported 

to limit Congress in exercise of its Article I war powers). For a contemporary discussion of 

the constitutional issue, see Quincy Wright, Treaties and the Constitutional Separation of 

Powers in the United States, 12 Am. J. Int'l L. 64, 65-85 (1918) (responding implicitly to 

Senator Lodge's arguments); see also Quincy Wright, The Control of American Foreign




pretive instrument-his use of the maxim is entitled to no less respect 

than Professor Tribe's. 

The point, however, is not to enter the fray on behalf of either of 

these approaches. From a purely textual perspective, Jefferson's reading 

is no more compelling than Professor Tribe's, and vice versa. 

What bears emphasis is Professor Tribe's failure even to consider 

Jefferson's mirror image reading and its implications for the exclusivist 

view. In making his categorical expressio unius argument, 

Professor Tribe just assumes that the maxim points his way. But when 

we expand the interpretive horizon, the self-defeating character of the 

argument becomes evident. 90 And it gives us powerful reasons for 

giving a skeptical greeting to arguments rooted in the kind of rigid 

interpretive rules he now seems to favor. 

Nor do matters improve when we descend to the details of Professor 

Tribe's expressio unius argument. He cites Alexander 

Hamilton in The Federalist as a strong proponent of the interpretive 

canon. 91 But he neglects to mention that in the very number on which 

he relies, Hamilton expressly denies that expressio unius applies to the 

construction of the Constitution. 92 On the contrary, Hamilton's careful 

discussion makes clear that it is only the underlying common sense 

inferences, not the maxim per se, that have any application to the 

Constitution. And his illustrations reveal the proper scope of common 

sense reasoning: When a nonexclusive reading would render a 

provision mere surplusage, the inference may properly be drawn. 93 

Thus, Article I's enumeration of Congress's powers and Article 1I1's 

Relations, § 54, at 97-98 (1922) [hereinafter Wright, Control] (arguing that treaty may constitutionally 

limit Congress's discretion in future exercise of power to declare war). 

90 This is not to claim that the two views cannot be harmonized. Indeed, Jefferson 

might well have thought that the expressio unius inference was properly applied to support 

both the exclusivity of the treaty power and its limitation to matters not falling within 

Congress's legislative authority. I assume, however, that Professor Tribe would not find 

this an appealing solution to the textual dilemma posed by Jefferson's argument. He supports 

application of expressio unius in the one case (i.e. to the Treaty Clause) but not in the 

other (i.e. to Congress's foreign affairs powers). 

91 See Tribe, Taking Text, supra note 3, at 1242-43 (citing The Federalist No. 83, at 496 

(Alexander Hamilton) (Clinton Rossiter ed., 1961)). 

92 See The Federalist No. 83, at 497 (Alexander Hamilton) (Clinton Rossiter ed., 1961). 

Hamilton wrote: "Even if these maxims had a precise technical sense, corresponding with 

the ideas of those who employ them upon the present occasion, which, however, is not the 

case, they would still be inapplicable to a constitution of government." Id. 

93 Hamilton is not alone among the early greats to hold this view. Chief Justice 

Marshall, for example, made the same point in Marbury v. Madison, 5 U.S. (1 Cranch) 137, 

174-75 (1803) (reasoning that "in this case, a negative or exclusive sense must be given to 

[the words of the Original Jurisdiction Clause], or they have no operation at all"), and then 

again in Cohens v. Virginia, 19 U.S. (6 Wheat.) 264,399-401 (1821) (explaining that arpressio 

unius was properly applied in Marbury because "otherwise, the [Original Jurisdiction 

Clause] would have no meaning whatever"). For further discussion, see infra Part IIIE.



enumeration of the judicial power are exclusive because any other 

reading would leave them without any operation at all. 94 In this respect, 

Hamilton's view accords with classical formulations of the expressio 

unius canon which emphasize that it "is not presumed that the 

legislature intended any part of a statute to be without meaning." 95 

It should be immediately evident that this analysis does not apply 

to the Treaty Clause. As even Professor Tribe concedes, the nonexclusive 

reading does not render the clause a nullity but leaves the 

President with a powerful option: He can bypass "the fires of bicameral 

approval" 96 and seek approval for his agreement through an extraordinary 

single house procedure, albeit one requiring a 

supermajority vote, or he can seek approval by simple majorities in 

both houses. 97 From a purely textual perspective, it is by no means 

clear that the President will always, or even more often, prefer the 

two-house route to the one-house advice and consent procedure. 98 

Thus, the exclusive reading cannot rely upon the (relatively) more 

persuasive inference Hamilton was able to draw in the case of Articles 

I and III. 

This is not to suggest that there is no force to the inference in 

favor of exclusivity in the case of the Treaty Clause, only that it is far 

weaker because it must rest upon an attenuated, less persuasive, chain 

of inferences: The Framers were explicit in granting the treaty power 

94 As Hamilton explains in regard to congressional powers: "This specification of particulars 

evidently excludes all pretension to a general legislative authority, because an affirmative 

grant of special powers would be absurd as well as useless if a general authority 

was intended." The Federalist No. 83, supra note 92, at 497. Likewise as to judicial power: 

"The expression of those cases marks the precise limits beyond which the federal courts 

cannot extend their jurisdiction, because the objects of their cognizance being enumerated, 

the specification would be nugatory if it did not exclude all ideas of more extensive authority." 

Id. See also his discussion in The Federalist No. 32, at 199-200 (Alexander Hamilton) 

(Clinton Rossiter ed., 1961), in which he argued that the power to tax domestic articles 

must be interpreted as being nonexclusive as between the federal and state governments to 

avoid rendering Article I, Section 10, Clause 2's prohibition on state taxation of imports 

and exports surplusage. 

95 Sutherland, supra note 87, at 412. For criticisms of this common sense view, consider 

Posner, supra note 86, at 281 (rejecting "no surplusage" canon as resting on mistaken assumption 

of "legislative omniscience"). 

96 Tribe, Comment, supra note 15, at 75. 

97 See Tribe, Taking Text, supra note 3, at 1250 n.98 (conceding this point). 

98 Witness, for example, the resistance of the early nineteenth-century executive to efforts 

by the House to end the practice of making treaties with Indian tribes and to require 

congressional approval of Indian agreements instead. See 2 Asher C. Hinds, Hind's Precedents 

of the House of Representatives of the United States § 1534 (1907) (recounting 

President Jackson's opposition to House participation in negotiation of Indian treaties); 

Jefferson, supra note 47, § 52, at 300 (noting that "in earlier times, [the Indian treaty] prerogative 

had been jealously guarded by the Executive"). For a discussion of the history of 

Indian agreement-making, see infra note 241.



1819 

to the President and the Senate in Article II; this suggests that they 

believed it to be an important power, which in turn suggests that they 

would have been equally explicit had they intended to vest any comparable 

power in Congress as a whole in Article I; the grant to Congress 

of the power to make all laws that are necessary and proper to 

carrying its enumerated powers into execution does not satisfy this 

standard; hence, it is not to be construed as a grant of power to Congress 

to authorize the President to conclude agreements on subjects 

within its enumerated powers. Perhaps this chain of reasoning, or 

some suitably refined alternative, accurately describes the Framers' intent. 

But, since, according to Professor Tribe, we may not consult the 

historical evidence, the inferences are highly speculative. It seems 

equally plausible, for instance, to think that the Framers were explicit 

about senatorial advice and consent because they were creating an 

extraordinary procedure that could not have been inferred from any 

other grant in the text, and because they wanted to make clear that 

the President would not have the power to make treaties unchecked 

by legislative scrutiny. 99 In the case of congressional approvals, in 

contrast, they could rest upon the implied powers doctrine that they 

explicitly wrote into the text in the Necessary and Proper Clause. 

Given the weakness of the exclusivity inference, expressio unius 

hardly provides the basis for a confident plain-meaning construction 

of the text. The contrast with a provision that would have no meaningful 

application in the absence of an exclusive reading is glaringly 

evident. 00 

Professor Tribe nevertheless repeatedly invokes expressio unius 

in support of his position. 01 After discussing Hamilton's view, he asserts 

broadly that the maxim "applies to provisions of the Constitution 

that both create entities and describe the powers those entities may 

wield."' 

°2 

Of course, any reliance on Hamilton for this claim would 

be fanciful, for, as we have seen, Hamilton never expressed anything 

remotely like this sweeping endorsement of exclusivity. 103 He rested 

instead on the more sensible and narrow proposition that an exclusive 

99 To the extent that the Framers believed that there was any ambiguity about whether 

approving agreements is properly regarded as a legislative or an executive function, the 

latter concern would have been more pressing. For discussion of the view of a number of 

key Framers that the power to approve agreements is properly legislative in character, see 

infra notes 247-62 and accompanying text. 

100 See supra note 67 and accompanying text. 

101 See Tribe, Taking Text, supra note 3, at 1241-43, 1269-71, 1273. 

102 Id. at 1243. 

103 See supra notes 93-94 and accompanying text.



[Vol. 73:1791 

interpretation may be proper when necessary to avoid rendering a 

provision in the text nugatory. 1 ' 4 

In any case, Professor Tribe's characterization of the expressio 

unius maxim is itself ambiguous in an important respect. Because of 

his reference to Hamilton's argument, one might suspect that he is 

endorsing a position that is at least parallel to Hamilton's. Thus, the 

principle might be thought to mean that the institutions created in 

Articles I, II, and III only have the powers specified in those articles: 

Congress is limited to the powers enumerated in Article I, the 

President to those in Article II, and the courts to those in Article III. 

On this reading, Professor Tribe presumably would argue that since no 

power to approve agreements is expressly specified in Article I, Congress 

has no such power. Alternatively, the quoted passage might be 

thought to mean that all of the powers enumerated in the articles creating 

those institutions should be construed as exclusively vested in 

the branch to which they are granted. So, all the powers granted to 

Congress in Article I are exclusively vested in it and cannot be exercised 

by any other branch, and so on for the other branches. On this 

reading, the argument presumably would be that Article II's grant of 

the treaty power to the President, with the participation of the Senate, 

is exclusive of any concurrent power in Congress under Article I. 

Initially, it is unclear why Professor Tribe might think either of 

these versions of the expressio unius principle advance the argument 

for an exclusive reading. 05 On a conceptual level, both are stymied 

104 In a letter, Professor Tribe indicates that the view stated in the passage quoted in the 

text is solely his own and that he did not intend to attribute it to Hamilton. See Tribe 

Letter to Golove of Apr. 30, 1998, supra note 80, at 3-4 (commenting, at my request, on 

this Article after its acceptance for publication). Without questioning this assertion, I note 

that there are strong textual reasons for interpreting the quoted passage otherwise. It is 

the final sentence in a paragraph devoted to discussion of Hamilton's view, and although 

that sentence never mentions Hamilton explicitly, the previous three sentences, along with 

an extended block quote from The Federalist No. 83, do so expressly, leaving the reader 

with a strong impression that Hamilton's discussion supports the principle asserted in the 

concluding sentence. See Tribe, Taking Text, supra note 3, at 1241-43. This effect is dramatically 

heightened by Professor Tribe's description of Hamilton's position as supporting 

the application of expressio unius "to provisions enumerating the limited powers of Congress 

and the limited jurisdiction of the federal courts." Id. at 1242 (citing The Federalist 

No. 83, supra note 92, at 496-97). While strictly accurate, this statement omits Hamilton's 

reasons for holding these views. Without such an explanation, not only is it a misleading 

description of Hamilton's position, but it also creates the impression that Hamilton's view 

supports Professor Tribe's immediately subsequent conclusion that expressio unitts "applies 

to provisions of the Constitution that both create entities and describe the powers 

those entities may wield." Id. at 1243. The important point, however, is that Professor 

Tribe and I are in agreement that nothing in Hamilton's discussion in The Federalist supports 

the broad view expressed in the passage quoted in the text. 

105 Indeed, in his letter, Professor Tribe indicated that he actually meant to endorse 

neither of these views and in fact supports only a narrower principle of exclusivity. See




by the Necessary and Proper Clause. Take the first alternative. Even 

assuming that each of the branches is strictly limited to the powers 

enumerated in the text, we still must determine whether the Necessary 

and Proper Clause, on which the nonexclusive reading principally 

rests, grants Congress the power to approve agreements. That clause 

is undeniably a part of Article I and an express grant of legislative 

authority. Likewise, the second version rests on the assumption that 

Tribe Letter to Golove of Apr. 30, 1998, supra note 80, at 1-3. According to Professor 

Tribe, what he meant to rest on is the more limited proposition "that constitutional provisions 

specifying how law is to be made should be presumed to be setting forth exclusive 

means of lawmaking." Id. at 1-2 (citing Tribe, Taking Text, supra note 3, at 1244). 

Notwithstanding this subjective intention, however, Professor Tribe's text virtually compels 

the far broader construction I have suggested. The quoted passage culminates his most 

sustained and explicit discussion of the maxim, and I am uncertain what other construction 

it, or a number of other passages, could be given. They all seem to express unequivocally 

the view that expressio unius should be applied generally to architectural provisions of the 

Constitution-provisions "that both create entities and describe the powers those entities 

may wield." Tribe, Taking Text, supra note 3, at 1243; see also id. at 1242 (attributing to 

Hamilton with apparent approval view that expressio inius is "properly applied to provisions 

enumerating the limited powers of Congress and the limited jurisdiction of the federal 

courts"); id. at 1246 (implicitly applying expressio unius to attack Professor Amar's 

claim that explicit, "architectural" provision granting immunity from civil arrest to members 

of Congress might be read to invite similar constitutionally-based immunity for 

President, and noting defects in Professor Ackerman's and Professor Amar's treatment of 

"architecture-defining, power-conferring provisions of the Constitution as merely suggestive"); 

id. at 1273 (applying expressio unius to Appointments Clause and arguing that 

Treaty Clause should be similarly construed); id. at 1274 n.181 (arguing that it is "a wiser 

course in constitutional interpretation to begin with the presumption that those provisions 

of the Constitution that call into being the very architecture of our government provide 

specific and exclusive instructions, not mere options"). 

Professor Tribe's apparent support for strict application of the expressio unius canon 

is further suggested by other arguments he presses at various points. See, e.g., id. at 1269- 

70 (arguing for application of expressio unius to provisions of Article I enumerating various 

consent-giving powers of Congress, thereby rendering them exclusive set); id. at 1271 

(arguing for application of expressio unius to Compact Clause, thereby rendering Congress's 

power to approve state agreements and compacts exclusive of any other congressional 

power over agreement-approving); id. at 1275-76 (arguing implicitly for application 

of expressio unius to Original Jurisdiction Clause and approving Chief Justice Marshall's 

controversial application of canon in Marbury v. Madison, 5 U.S. (1 Cranch) 137 (1803), 

discussed in Part uEI.E, infra). To be sure, Professor Tribe's article also made the narrower 

argument he now articulates in his letter. See, e.g., id. at 1244 (arguing that "the most 

plausible way of reading the Constitution... would be to read as exclusive those provisions 

that specify how elements of the supreme law of the land are to be adopted"). But in 

his original text, that argument was additional to, and separate from, the broader erpressio 

unius argument that he also articulated at numerous points in his discussion. It was cast, 

moreover, as a structural argument resting principally upon federalism grounds. As so 

understood, I criticize this view at length in Part III.C.1, infra. See also infra note 126 

(noting Professor Tribe's emphasis on Framers' concern for state sovereignty). Because 

the broader expressio argument is so strongly suggested by his text, I continue to attribute 

it to Professor Tribe. In doing so, however, I do not mean to question his characterization 

of his subjective intention. This confusion over the "plain meaning" of Professor Tribe's 

text is itself an ironic commentary on the very hazards of textualism I hope to illustrate.



[Vol. 73:1791 

the power in question is granted only to one branch, but where there 

is a textual basis for finding an overlapping grant in both Articles I 

and II, the principle would seem to have little or no application. 

Overlapping powers require harmonization, not categorical division. 

Before drawing Professor Tribe's expressio unius inference, then, one 

must first interpret the grants in Article I to determine whether they 

include a power to authorize agreements. To be sure, in the interpretive 

enterprise, the more modest expressio unius argument that I have 

endorsed may indeed be relevant, but Professor Tribe has given us no 

reason to suppose that his strict version of the canon adds to the 

analysis. 

In any case, application of either version of the principle would 

yield a government unrecognizable to We the People, for it fundamentally 

contradicts the way the Court and the political branches have 

construed the text over the long course of our constitutional history. 

Take the first interpretation-that each branch is strictly limited to the 

powers enumerated in the article that creates that branch. Hamilton 

himself expressly rejected this reading of Article II in his famous argument 

defending Washington's controversial 1793 declaration of neutrality 

in the war between France and Great Britain. He claimed that 

the explicit enumerations of executive power in Article II, in contrast 

to those in Articles I and III, were merely exemplary of the 

President's powers, 10 6 and that Article II's vesting of the executive 

power "in a President of the United States of America" 1 7 constituted 

a lodging of any and all powers that are executive in character. 108 

There is surely some irony in Hamilton's argument, given his earlier 

endorsement of an exclusive reading of Articles I and III to avoid 

rendering their provisions nugatory. But this contrast only underscores 

the importance of Hamilton's caution about the dangers of invoking 

the expressio unius principle, even when strong linguistic 

arguments can be made to support it. It is doubly ironic that Professor 

Tribe himself explicitly endorses Hamilton's nonexclusive reading 

of Article II, but never pauses to consider the far stronger grounds for 

applying expressio unius to Article II than to the Treaty Clause. 109 

106 See Alexander Hamilton, Pacificus No. 1 (June 29, 1793), reprinted in 15 The Papers 

of Alexander Hamilton 33, 39 (Harold C. Syrett ed., 1969) [hereinafter Pacificus No. 1]. 

Writing as Pacificus in the celebrated Pacificus-Helvidius debate with Madison, Hamilton 

claimed that the enumerated powers were "intended by way of greater caution, to specify 

.and regulate the principal articles implied in the definition of executive power; leaving the 

rest to flow from the general grant of power." Id. at 39. 

107 U.S. Const. art. II, § 1. 

108 See Pacificus No. 1, supra note 106, at 39. 

109 See Tribe, Taking Text, supra note 3, at 1269 & n.165 (explicitly endorsing 

Hamilton's view); see also Tribe, Constitutional Law, supra note 11, § 4-2, at 210-11 & 210


December 1998] AGAINST FREE-FOPM FORAL4LISM 

1823 

Now, consider the application of this version of the principle to 

the powers of Congress enumerated in Article I.110 Pace Hamilton, 

n.1 (same). In contrast to Professor Tribe, many still find Hamilton's position unsettling 

(myself among them), see infra note 342 and accompanying text, but it is difficult to deny 

its force without casting doubt upon important aspects of the modem presidency. See 

Henkin, supra note 9, at 39-41 (describing difficulty of accounting for President's expansive 

power over foreign affairs on basis of enumerated powers alone). 

Like Hamilton, Professor Tribe rests his view largely on differences in the wording of 

the vesting clauses of Articles I and HI. Article I provides that "[a]ll legislative Powers 

herein granted shall be vested in a Congress of the United States," U.S. Const. art. I, § 1 

(emphasis added), whereas Article I begins: "The executive Power shall be vested in a 

President of the United States of America," id. art. II, § 1. On this basis, Professor Tribe 

argues that the Framers intended thereby to limit Congress to its enumerated powers (i.e., 

the powers "herein granted") but to allow the President to exercise full executive powers, 

the express enumerations in Article HI apparently being merely exemplary of, or in some 

manner limiting, the more general grant. See Tribe, Constitutional Law, supra note 11, § 4- 

2, at 210-11 & 210 n.1; Tribe, Taking Text, supra note 3, at 1269. There are numerous and 

obvious textual difficulties with the argument. Not the least of these is its inability to explain 

persuasively why the Framers would have burdened Article II with a list of enumerated 

powers that would obviously have been subsumed under the Hamiltonian Executive 

Power Clause-such as, for example, the commander-in-chief power, see U.S. Const. art. 

11, § 2, cl. 1; the power to require written opinions of the heads of departments, see id.; the 

pardon power, see id.; and the duty to faithfully execute the laws, see id. art. I1, § 3. Curiously, 

given his reliance on the expressio wijus maxim, Professor Tribe never confronts this 

powerful expressio unius objection. Equally important, Article III, like Article HI but unlike 

Article I, unreservedly vests the "judicial Power of the United States ... in one 

supreme Court." Id. art. IlI, § 1. Yet, it is settled that the judicial power extends only to 

the jurisdictional grants enumerated in Article I, Section 2. See, e.g., Sheldon v. Sill, 49 

U.S. (8 How.) 441, 449 (1850) (holding that Congress may subtract from, but not add to, 

jurisdiction granted to lower federal courts in Article ). But see Kansas v. Colorado, 206 

U.S. 46, 81-84 (1907), in which Justice Brewer argued in obscure dicta for a Hamiltonian 

reading of Article III on the same textual grounds Hamilton advanced in relation to Article 

11-an opinion no doubt meriting in this respect the appellation "derelict on the waters 

of the law." Ironically, the wording of all three articles was the same until late in the 

Convention when Gouveneur Morris, charged with making purely stylistic revisions, surreptitiously 

slipped the qualifying language into Article L Some have thought, with substantial 

justification, that he did so in an effort to work substantive changes without calling 

the Convention's attention to his handiwork. See Abraham D. Sofaer, War, Foreign Affairs 

and Constitutional Power The Origins 37 (1976). In any event, Professor Tribe's 

acceptance of the Hamiltonian reading simply suggests how weak the expressio unius principle 

is even in his constitutional universe. If the compelling case for its application 

presented by Article II can be overcome by the admittedly weak textual case which both 

he and Hamilton accept as sufficient, then surely the far weaker erpressio unius inference 

from the Treaty Clause-which cannot take strength from the imperative to avoid rendering 

constitutional language nugatory--can be overcome by the considerably more persuasive 

textual basis for authorizing agreements found in the Necessary and Proper Clause. 

110 If powers are expressly enumerated in other parts of the Constitution, as they in fact 

are in a number of cases, Congress can of course exercise them even if we were to give 

Article I an exclusive reading. Consider, for example, the powers listed in Article IV, 

which include certain authority over the implementation of the Full Faith and Credit 

Clause, see U.S. Const. art. IV, § 1; the power to admit new states, see id. art. IV, § 3, cl. 1; 

the power to dispose of and make all needful rules and regulations respecting the territory 

and property of the United States, see id. art. IV, § 3, cl. 2; and whatever powers are 

implied in the obligation to guarantee each state a republican form of government, see id.



much judicial rhetoric, and even the Tenth Amendment,' the claim 

that Congress is strictly limited to the enumerated powers is simply as 

a categorical matter untrue. At times, the Court has slipped in new 

powers Congress wished to exercise under the Necessary and Proper 

Clause door." 2 At others, it has largely eschewed reliance on the text 

and simply found important powers to belong to Congress because 

they are inherent in the concepts of nationhood and sovereignty under 

international law. In United States v. Curtiss-Wright Export Corp.,113 

the Court claimed that the federal government's powers of external 

sovereignty do not depend upon the affirmative grants in the Constitution 

but are vested in it "as necessary concomitants of nationality.""1 

4 One does not have to accept the full implications of Justice 

Sutherland's controversial theory, however, to notice that throughout 

our history the Supreme Court has affirmed congressional powers nowhere 

to be found within the four comers of the text but premised on 

the inherent sovereignty of the United States. 115 

art. IV, § 4. The powers enumerated in Article I would be exclusive of any powers not 

included in that article or in any other part of the Constitution. 

111 See id. amend. X. The Tenth Amendment directs that the "powers not delegated to 

the United States by the Constitution, nor prohibited by it to the States, are reserved to the 

States respectively, or to the people." Id. 

112 Under the Necessary and Proper Clause, the Court has permitted Congress to charter 

banks and other corporations, see McCulloch v. Maryland, 17 U.S. (4 Wheat.) 316 

(1819) (upholding Congress's power to charter second Bank of United States); The Pacific 

Railroad Removal Cases, 115 U.S. 1, 14-15 (1885) (upholding Congress's power to charter 

railroad corporations); to make paper money a legal tender, see The Legal Tender Cases, 

79 U.S. (12 Wall.) 457 (1870) (upholding implicitly Congress's power to make treasury 

notes legal tender in payment of all debts and obligations); and to undertake investigations 

with the power to issue subpoenas and punish contempts, see McGrain v. Daugherty, 273 

U.S. 135 (1927) (upholding Congress's investigatory power, including its power to compel 

testimony by subpoena); Anderson v. Dunn, 19 U.S. (6 Wheat.) 204 (1821) (upholding 

Congress's power to hold nonmembers in contempt of Congress for attempting to bribe 

member). 

113 299 U.S. 304 (1936). 

114 Id. at 318. 

115 Perhaps the most famous instance was the Court's opinion in the Chinese Exclusion 

Case, 130 U.S. 581, 603-04 (1889) (upholding power to exclude aliens on ground that it is 

an incident of every independent nation). But there are many others. See, e.g., Perez v. 

Brownell, 356 U.S. 44, 57-60 (1958) (upholding power of Congress to withdraw citizenship 

of citizens who vote in foreign elections), overruled on other grounds in Afroyim v. Rusk, 

387 U.S. 253 (1967); Blackmer v. United States, 284 U.S. 421, 436-38 (1932) (upholding 

power to compel U.S. citizens to return from abroad to testify in criminal proceedings); 

Fong Yue Ting v. United States, 149 U.S. 698, 705-14 (1893) (finding power to exclude or 

expel aliens inherent in sovereignty); Jones v. United States, 137 U.S. 202, 212 (1890) (finding 

authority to acquire territory by discovery and occupation and exercise jurisdiction 

over it in rights of nations under international law); Late Corp. of the Church of Jesus 

Christ of Latter-Day Saints v. United States, 136 U.S. 1, 42 (1890) (upholding power to 

acquire territory by conquest or treaty); American Ins. Co. v. Canter, 26 U.S. (1 Pet.) 511, 

542-43 (1828) (Marshall, C.J.) (same). The latter three cases, among others, resolved any




1825 

doubts about whether the Constitution permitted the United States to acquire new territory, 

a question that had been opened by President Jefferson's ruminations over the constitutionality 

of the Louisiana Purchase. See 1 Westel Woodbury Willoughby, The 

Constitutional Law of the United States § 232, at 410-12 (2d ed. 1929) (discussing 

Jefferson's views). For an excellent discussion of the many cases in which courts have 

upheld congressional powers on the basis of American sovereignty and actual practices of 

Congress that have never received judicial sanction but are best explained on that basis, 

see Henkin, supra note 9, at 21, 70-72 & 361-65 nn.29-49, 366 n.53. Among the former are 

those described above as well as the power to provide for extradition in the absence of a 

treaty, to require aliens to register, and to take away the citizenship of women who marry 

aliens (now archaic). Among the latter are the power to regulate foreign diplomatic activities 

in the United States, to freeze the assets of foreign states, to assert national sovereignty 

in airspace or in special zones at sea, to pass laws protecting the civil rights of aliens, and to 

enact a comprehensive criminal code applying to citizens abroad. See id. Professor Tribe 

has recognized that some of Congress's powers cannot be derived from any particular 

grant in the text. See Tribe, Constitutional Law, supra note 11, § 5-3, at 304-05 (noting that 

"certain peripheral congressional powers" result from "the aggregate of national powers 

and the 'nature of a political society"'). 

An advocate of this version of Professor Tribe's original principle will also have difficulty 

explaining the inconsistency betveen a strong affirmation of expressio unis and another 

key aspect of our constitutional structure. Consider how after enumerating 

Congress's powers, Article I then devotes a section to specifying which of those powers are 

to be exclusive of the states. See U.S. Const. art. I, § 10. Section 10 lists a number of 

powers that are specifically denied the states, and most of these are direct cognates of the 

powers given to Congress in Section 8. Compare, e.g., id. art. I, § 8, cl. 5 (granting power to 

coin money and regulate value thereof), with id. art. I, § 10, c. 1 (prohibiting states from 

coining money). This structure, in turn, reflects the fact that before the Constitution the 

states were largely independent, and with the adoption of the Constitution their sovereignty 

was left intact except insofar as the document indicates otherwise. See, e.g., Sturges 

v. Crowninshield, 17 U.S. (4 Wheat.) 122, 192-93 (1819) (Marshall, C.J.) (describing same 

in holding bankruptcy power nonexclusive). Thus, there are strong grounds for construing 

those Section 8 powers for which there is no Section 10 cognate exclusion as nonexclusive, 

as the Supreme Court has often indicated. See, e.g., id. at 193; see also The Federalist No. 

32, supra note 94, at 199-201 (advocating same principle). Still, the determination as to 

whether a particular grant is exclusive of state authority will depend upon a careful examination 

of its nature and purposes. As Marshall put it: "Whenever the terms in which a 

power is granted to Congress, or the nature of the power, require that it should be exercised 

exclusively by Congress, the subject is as completely taken from the State Legislatures, 

as if they had been expressly forbidden to act on it." Sturges, 17 U.S. (4 Wheat.) at 

193; see also Holmes v. Jennison, 39 U.S. (14 Pet.) 540,574 (1840) (opinion of Taney, CJ.) 

(concurring in this view); The Federalist No. 32, supra note 94, at 220 (same). Examples of 

Section 8 powers that are exclusive vis-a-vis the states notwithstanding their noninclusion 

in Section 10 are the power to borrow money on the credit of the United States, see U.S. 

Const. art. I, § 8, cl. 2; to establish post offices and post roads, see id. art. I, § 8, cl. 7; and to 

constitute tribunals inferior to the Supreme Court, see id. art. I, § 8, cl. 9. There are others. 

See supra note 105. Even though the states are expressly denied a number of foreign 

affairs powers in Section 10, in Zschernig v. Miller, 389 U.S. 429 (1968), the Court found a 

further unenumerated limit on the states that barred Oregon from denying inheritance to 

an East German citizen in accordance with its policy to deny estates to citizens of the 

former German Democratic Republic. Under a strict expressio unius interpretive regime, 

it would seem that all of the powers enumerated in Section 8 for which there is no Section 

10 cognate ought to be deemed nonexclusive. That this is simply not so merely reflects the 

fact that we do not live in a strict expressio unius regime, Professor Tribe's protestations 

notwithstanding.



[Vol. 73:1791 

The second version of the expressio unius principle suggested by 

Professor Tribe's text fares no better-indeed, far worse-as a description 

of our constitutional heritage. Under this reading, the claim 

is that a power granted to one branch ought to be read as exclusive of 

any overlapping power in the other branches. If this argument applies 

to individual grants of power, a fortiori it would apply to the broad 

categorical grants of Articles I, II, and III-legislative power to the 

Congress, executive power to the President, and judicial power to the 

Courts. 116 Yet, even on this macro plane, the thesis ill fits our practices. 

Consider the so-called independent agencies. Acting under its 

Article I authority and the Necessary and Proper Clause, Congress is 

free to create a shadow executive branch substantially independent of 

presidential control by prohibiting the President from removing officers 

of independent agencies except for cause. 117 Or, consider the 

Court's longstanding endorsement of Article I courts. Acting under 

Article I and the Necessary and Proper Clause, Congress may vest 

116 This reading is supported by Professor Tribe's sweeping but unsupported assertion 

that "those provisions of the Constitution that call into being the very architecture of our 

government provide specific and exclusive instructions, not mere options." Tibe, Taking 

Text, supra note 3, at 1274 n.181. If this were true, how could we explain, as discussed 

below, independent executive agencies, Article I courts, federal common law, and the 

President's legislative powers in foreign affairs See infra notes 117-18 and accompanying 

text. Here, too, I suspect that Professor Tribe would not now endorse the apparent implications 

of his text. 

117 Humphrey's Executor v. United States, 295 U.S. 602 (1935), is the leading case, from 

which the Supreme Court has never substantially retreated. See id. at 629-31 (affirming 

power of Congress to restrict presidential removals of officers of quasilegislative agencies); 

Wiener v. United States, 357 U.S. 349, 353-56 (1958) (discussing and following Humphrey's 

Executor in context of quasijudicial agency); Morrison v. Olson, 487 U.S. 654, 687-91 

(1988) (same). In Morrison, the Court upheld the power of Congress to restrict the removal 

of even some core executive officers, in that case independent counsel appointed 

pursuant to the Ethics in Government Act. For further discussion of the Appointments 

Clause, see infra Part III.A.2. Despite the unambiguous grant of the executive power to 

the President, Professor Tribe himself agrees that "there is nothing to stop the legislature 

from vesting executive authority in officers substantially independent of the White House." 

Tribe, Constitutional Law, supra note 11, § 4-10, at 253. For good discussions of the issues 

posed by the independent agencies, see Peter M. Shane, Independent Policymaking and 

Presidential Power: A Constitutional Analysis, 57 Geo. Wash. L. Rev. 596 (1989); Peter L. 

Strauss, The Place of Agencies in Government: Separation of Powers and the Fourth 

Branch, 84 Colum. L. Rev. 573 (1984). For recent debates about the Framers' intentions, 

compare Steven G. Calabresi & Saikrishna B. Prakash, The President's Power to Execute 

the Laws, 104 Yale Li. 541, 642-45 (1994) (arguing that members of First Congress conceived 

of executive as unitary branch and agreed that President either had power to remove 

all inferior executive officers or none), with Martin S. Flaherty, The Most Dangerous 

Branch, 105 Yale L.J. 1725, 1788-92 (1995) (disputing idea that Framers shared any clear 

conception of scope of executive branch and hence of Congress's power to create independent 

agencies), and Lawrence Lessig & Cass R. Sunstein, The President and the Administration, 

94 Colum. L. Rev. 1, 106-08 (1994) (arguing that insulating offices such as Federal 

Reserve Board from presidential authority is consistent with Framers' intended scheme).




judicial business that could have been assigned to the federal courts 

under Article Ill to tribunals presided over by judges not enjoying the 

special protections afforded Article Ill judges. 118 

118 See U.S. Const. art. II, § 1. Federal judges, of course, are entitled to life tenure and 

salary protection. See id. The leading contemporary case is Northern Pipeline Constr. Co. 

v. Marathon Pipe Line Co., 458 U.S. 50, 69-70, 83-84 (1982) (striking down bankruptcy 

courts but affirming wide area for Article I courts in cases arising under federal law). The 

earlier cases begin with Chief Justice Marshall's opinion in Canter, 26 U.S. (1 Pet.) at 545 

(upholding congressional power to assign territorial courts jurisdiction over admiralty 

cases), and have continued unimpeded since. See, e.g., Murray's Lessee v. Hoboken Land 

& Improvement Co., 59 U.S. (18 How.) 272, 275, 282-85 (1856) (distinguishing between 

matters that are inherently judicial and those raising questions of public right that may be 

assigned to non-Article III tribunals); Ex parte Bakelite Corp., 279 U.S. 438,458-59 (1929) 

(holding that Court of Customs Appeals was Article I court); O'Donoghue v. United 

States, 289 U.S. 516, 551 (1933) (holding courts of District of Columbia to be Article LII 

courts); Williams v. United States, 289 U.S. 553,581 (1933) (holding Court of Claims to be 

Article I court); Glidden v. Zdanok, 370 U.S. 530, 541-43 (1962) (holding Court of Claims 

and Court of Customs and Patent Appeals to be Article Im courts). The Court's decisions 

may be "landmarks on a judicial 'darkling plain' where ignorant armies have clashed by 

night" Northern Pipeline, 458 U.S. at 91 (Relnquist, J., concurring) (characterizing Justice 

White's view of Court's past precedents). Nevertheless, even in the midst of the epochal 

battle between Justices Brennan and White over how to preserve the soul of Article I1, 

neither called into question the very substantial power of Congress to give cases arising 

under federal law over to Article I tribunals and administrative agencies. Compare id. at 

69-70, 83-84 (Brennan, J.), with id. at 113-15 (White, J., dissenting). Justice Brennan, 

speaking for a plurality, sought to confine the nonexclusivity of Article III to three categories: 

territorial courts, courts-martial, and courts or agencies adjudicating claims falling 

under the vaguely drawn "public rights" doctrine (i.e., cases between the government and 

others which could have been resolved solely by the executive without any judicial role). 

See id. at 63-70 (Brennan, J.). Justice White thought the CourVs past precedents could not 

be so easily cabined and preferred to recognize far more discretion in Congress to assign 

matters within the judicial power to non-Article III tribunals. See id. at 103-05, 114-15 

(White, J., dissenting) (arguing for loose balancing test under which Congress would be 

free to assign cases to non-Article III tribunals so long as Article I values are not substantially 

undermined). Both, however, sought to assure Congress that the administrative state 

as we know it was not under a constitutional cloud. See id. at 69-70, 83-84 (Brennan, J.) 

(affirming that Congress may provide that persons seeking to enforce particular statutorily 

created rights must do so before administrative tribunals); id. at 113-15 (White, J., dissenting) 

(concluding, in light of large body of precedent, that it was "too late" to deny constitutionality 

of administrative agencies). Even the careful limits Justice Brennan sought to 

establish in Northern Pipeline for state law actions, moreover, have not been strictly 

respected by the Court in subsequent decisions. See Commodity Futures Trading Commn 

v. Shor, 478 U.S. 833 (1986) (allowing non-Article I tribunal to adjudicate state law 

claims in certain cases); cf. Thomas v. Union Carbide Agric. Prods. Co., 473 U.S. 56S 

(1985) (expanding public rights exception). 

Furthermore, when we turn from legislative intrusions on the executive and judicial 

realm to judicial and executive trespasses on the legislative function, we continue to discover 

overlapping powers. The quasilegislative powers of the federal courts to promulgate 

federal common law are well known if still controversial. For recent examples, see Boyle v. 

United Techs. Corp., 487 U.S. 500, 511-13 (1988) (fashioning federal common law defense 

for federal contractors sued in state law products liability actions); Wesifall v. Erwin, 484 

U.S. 292, 295-98 (1988) (considering limits of the federal common law immunities of federal 

officials); United States v. Little Lake Misere Land Co., 412 U.S. 5S0, 592 (1973) (ap-



[Vol. 73:1791 

When we descend from the heights of these meta-examples to the 

level of individual powers, this version of the expressio unius principle 

is equally unsustainable. It is simply false that the powers granted to a 

branch are uniformly exclusively held by that branch. 119 Counterexamples 

are legion. Thus, the President has been allowed to invade a 

plying Clearfield Trust federal common law test, see Clearfield Trust Co. v. United States, 

318 U.S. 363, 366-67 (1943), to determine law applicable to contract suits brought by 

United States). The federal courts often create a kind of quasiconstitutional common law. 

The dormant Commerce Clause cases, among others, are probably best understood in this 

light. See Henry P. Monaghan, The Supreme Court, 1974 Term-Foreword: Constitutional 

Common Law, 89 Harv. L. Rev. 1, 17 (1975) (arguing in favor of this view). But see 

Thomas S. Schrock & Robert C. Welsh, Reconsidering the Constitutional Common Law, 

91 Harv. L. Rev. 1117, 1138-41 (1978) (rejecting common law account of Commerce Clause 

jurisprudence). In the foreign affairs context, the most important example is Banco Nacional 

de Cuba v. Sabbatino, 376 U.S. 398, 421-27 (1964) (establishing, as matter of federal 

common law, modem act of state doctrine). The separation of powers concerns raised by 

judicial lawmaking underlie the longstanding dispute over implied private rights of action, 

compare, e.g., Thompson v. Thompson, 484 U.S. 174 (1988) (refusing to imply right of 

action under Parental Kidnapping Prevention Act of 1980), with Cannon v. University of 

Chicago, 441 U.S. 677 (1979) (implying a right of action under Title IX of Education 

Amendments of 1972, and debating, in sharply divided decision, separation of powers issues 

posed by implied rights of action), and, to some degree, over the scope of Bivens 

actions, see, e.g., Schweiker v. Chilicky, 487 U.S. 412 (1988) (dividing over propriety of 

implying Bivens remedy when Congress has enacted elaborate remedial scheme on 

subject). 

Less well known, perhaps, are the President's legislative powers in foreign affairs, 

including, for example, his recognized power, in the absence of an act of Congress, to 

determine whether a foreign state will be accorded sovereign immunity by courts in the 

United States. See Republic of Mexico v. Hoffman, 324 U.S. 30, 36 (1945) (upholding 

executive authority to make binding suggestion of immunity); Ex parte Republic of Peru, 

318 U.S. 578, 588 (1943) (same). Professor Henkin provides an instructive discussion of 

the President's legislative powers in foreign affairs. See Henkin, supra note 9, at 54-61. 

For a detailed description with supporting citations of the legislative and judicial powers of 

the President in war zones and in occupied territory, see Clarence Arthur Berdahl, War 

Powers of the Executive in the United States 152-64 (1921); see also Madsen v. Kinsella, 

343 U.S. 341, 348 (1952) (acknowledging President's power to establish and regulate tribunals 

in territory under military occupation); Cross v. Harrison, 57 U.S. (16 How.) 164, 189- 

90 (1853) (affirming President's authority to form civil government, establish port regulations, 

and impose duties on imports and tonnage in conquered territory). 

119 That is the point, I take it, of Justice Jackson's famous tripartite division of executive 

activities in the Steel Seizure Case, Youngstown Sheet & Tube Co. v. Sawyer, 343 U.S. 579, 

635-38 (1952) (Jackson, J., concurring): "[T]here is a zone of twilight in which [the 

President] and Congress may have concurrent authority. .. ." Id. at 637. Nowhere is the 

overlap more extended than in foreign affairs. As Professor Henkin puts it: 

Jackson wrote in the Steel Seizure Case, which the Court treated as domestic, 

not as involving foreign affairs. Even in domestic matters, Jackson implies, 

there is a "twilight zone" clearly within the constitutional domain of Congress 

in which the President could also act. In foreign affairs, surely, where the 

President admittedly has large power, the fact that Congress can act does not, 

of itself, prove that the President could not; Presidents, we have seen, have 

acted unilaterally in foreign affairs matters which Congress might undoubtedly 

have regulated, where Congress had not in fact done so.




variety of Congress's express powers, 120 and Congress has returned 

Henkin, supra note 9, at 95 (footnote omitted); see also id. at 92-123 (describing some 

areas of overlap in foreign affairs). Specific examples are discussed infra notes 120-23, 131- 

35, 138, 197-209, 398-403, and accompanying text. 

Of course, exclusivity, in the constitutional context, lies more on a spectrum than in 

two opposite poles. The question is usually better stated as exclusivity as to what and as 

against whom, rather than exclusivity per se. Even the concepts of exclusivity and nonexclusivity 

have many variegated meanings depending upon context. The Appointments 

Clause, U.S. Const. art. II, § 2, cl. 2, is a case in point. Professor Tribe believes that it is 

common ground that the senatorial advice and consent procedure is exclusive as to other 

means of appointment in the case of principal officers. See Tribe, Taking Text, supra note 

3, at 1272. But even this much is not entirely so. Congress has long purported to insist that 

(some) appointees meet specified qualifications. See Henkin, supra note 9, at 122 & 4(10 

nn.104-05. More significant, although the Appointments Clause explicitly mandates the 

senatorial advice and consent procedure for "Ambassadors, other public Ministers and 

Consuls," U.S. Const. art. II, § 2, cl. 2, presidents, with the acquiescence of senates and 

congresses, have from the beginning disregarded this command for certain presidential 

appointees charged with carrying out foreign policy tasks on his behalf. As Corwin put it 

nearly a century ago: 

Such agents have been justified as "secret agents," yet neither their existence 

nor their mission is invariably secret. They have been called "private agents of 

the President," his "personal representatives," yet they have been sometimes 

commissioned under the great seal. They have been justified as organs of negotiation 

and so as springing from the Executive's power in negotiating treaties, 

yet this is also a normal function of our regular representatives. They 

have been considered as agents appointed for special occasions, but, as we 

have seen, the term "public ministers" of the Constitution is broad enough to 

include all categories of diplomatic agents.... In short, the only test which is 

generally available for distinguishing this kind of agents [sic] from the other 

kind is to be found in the method of their appointment. 

Edward S. Corwin, The President's Control of Foreign Relations 65 (1917) [hereinafter 

Corwin, President's Control]; see also Edward S. Corwin, The President: Office and Powers, 

1787-1984, at 236-37 (Randall W. Bland et al. eds., 5th rev. ed. 1984) [hereinafter 

Corwin, The President] (noting that beginning with George Washington presidents have 

appointed "secret agents" without advice and consent of Senate); Corwin, President's Control, 

supra, at 58-66 (same); Henkin, supra note 9, at 42 & 340 n.22 (same). See generally 

Henry Merritt Wriston, Executive Agents in American Foreign Relations (1929). For an 

early Attorney General opinion upholding the practice, see 7 Op. Att'y Gen. 186, 204-06, 

212-13 (1856). This practice, with its roots in President Washington's administration, ought 

to provide a clear warning against categorical thinking and rigid demarcations in the field 

of foreign affairs. In any case, it is not only the Appointments Clause's procedure for 

appointing principal officers that is nonexclusive in part. The same applies to the 

President's power to remove principal officers. While Congress may not participate directly 

in removals, it may in some cases limit the substantive grounds on which the 

President may remove. See, e.g., Morrison, 487 U.S. at 685-91 (holding that Congress has 

power to establish a "good cause" requirement for removal of independent counsel). In 

others, however, where it would undermine the President's ability to carry out his constitutional 

duties, it cannot. See id. (stating test). 

120 Consider, for example, Congress's power to make rules for the government and regulation 

of the land and naval forces. See U.S. Const. art. I, § 8, cl. 14. Despite the explicit 

grant to Congress, presidents have asserted, and the Court has held, that the President has 

inherent power as Commander-in-Chief to constitute courts-martial and impose other 

rules for governing military life. See Swaim v. United States, 165 U.S. 553, 557-58 (1897) 

(upholding President's power to constitute courts-martial); see also Kurtz v. Moffitt, 115



the favor, exercising fully or partially concurrent authority in a 

number of areas of enumerated presidential power. 21 Congress has 

also legislated in areas assigned to the judiciary.' 22 It thus seems that 

the expansive understanding of the expressio unius maxim suggested 

by Professor Tribe's text is utterly unfounded.'2 

U.S. 487, 503 (1885) (giving effect to regulations providing rewards for apprehension of 

deserters promulgated on President's own authority); United States v. Eliason, 41 U.S. (16 

Pet.) 291, 301 (1842) (stating that "power of the executive to establish rules and regulations 

for the government of the army, is undoubted," although in that case there was act of 

Congress); Berdahli, supra note 118, at 138-42 (stating that Commander-in-Chief Clause 

empowers President to convene and regulate courts-martial); Henkin, supra note 9, at 46 

(noting that presidents have claimed authority to make rules for government and regulation 

of land and naval forces based on commander-in-chief power). 

Likewise, under his powers as sole organ of the nation in conducting its foreign affairs 

and as Commander-in-Chief, the President has invaded the spheres of many other congressional 

powers, including even the power to regulate foreign commerce. See Henkin, supra 

note 9, at 41-46,54-61, 92-96, 103-05 (providing examples). Consider, for example, the case 

of foreign sovereign immunity, the international law doctrine that affords foreign sovereigns 

immunity from suit in United States courts. Until 1976, when Congress enacted the 

Foreign Sovereign Immunities Act, Pub. L. No. 94-583 § 4(a), 90 Stat. 2892 (1976) (codified 

as amended at 28 U.S.C. §§ 1602-1611 (1994)), the State Department could in any case 

involving a foreign sovereign submit a "suggestion" as to immunity that would bind the 

court. See Hoffman, 324 U.S. at 36 (upholding Executive Branch's power); Ex parte Republic 

of Peru, 318 U.S. at 589 (same). In 1976, however, Congress dethroned the Executive 

and, promulgating an elaborate regulatory scheme, asserted its right, presumably at 

least in part under its foreign commerce powers, to control the field. See Foreign Sovereign 

Immunities Act, 28 U.S.C. §§ 1602, 1611 (1994) (removing presidential authority to 

make suggestion of immunity and replacing it with objective standards for judicial application). 

Several presidents, without statutory authority, have permitted the landing of foreign 

submarine cables, see 22 Op. Att'y Gen. 13 (1898); Corwin, The President, supra note 

119, at 226-27, and the introduction of electricity from abroad, see 30 Op. Att'y Gen. 217 

(1913); Corwin, The President, supra note 119, at 227. Without statutory authority, President 

Wilson ordered the closure of a foreign radio station because of noncompliance with 

naval censorship regulations. See 30 Op. Att'y Gen. 291 (1914) (justifying closure of Marconi 

station under commander-in-chief power); Corwin, The President, supra note 119, at 

227 (noting same); Henkin, supra note 9, at 345 n.43 (same). 

Under his power to make unilateral executive agreements, the President has frequently 

done the same. See Henkin, supra note 9, at 43, 219, 225-26 (giving examples). 

The war powers of the two branches are also substantially overlapping, see infra note 209; 

so too may be the commander-in-chief power and the power to make rules for captures on 

land and water. See infra note 209; infra notes 197-98 and accompanying text. 

121 The pardon power is entirely nonexclusive, see infra notes 199-201 and accompanying 

text, while the power to require written opinions by the heads of departments seems 

largely, if not entirely, so, see infra notes 202-04 and accompanying text. For other examples, 

see infra note 209 and accompanying text. 

122 See the discussion of the admiralty and maritime power infra notes 131-34 and accompanying 

text. 

123 Perhaps an even more telling example is the immigration power. Not only has the 

Court long recognized it as an unenumerated power of Congress implied in the nation's 

sovereign status under international law, see supra note 115 and accompanying text, it has 

also found that the President has a partially concurrent implied power over the subject. 

See United States ex rel. Knauff v. Shaughnessy, 338 U.S. 537, 542-43 (1950) (holding that 

power to exclude aliens is "inherent in the executive department of the sovereign"). For




Fortunately though, it now appears that this conclusion is common 

ground and that Professor Tribe wishes to stand on what he believes 

is a less sweeping version of the expressio unius principle.u 4 As 

he explains in a recent communication, in this version of the argument 

not all provisions creating entities and describing their powers should 

be given an exclusive construction; only those specifying how elements 

of the supreme law of the land are to be adopted need be accorded 

this presumption.125 Professor Tribe made a similar argument 

in his original text, but in that formulation the argument appeared to 

be structural, resting chiefly on federalism considerations. 12 6 As so 

understood, I devote a separate section to criticizing that argument 

below. 2 7 But in his current articulation, he seems to have something 

more general in mind, implying that this principle can be unmoored 

from its federalism underpinnings and pressed directly as an expressio 

unius argument.'-m Unfortunately, though, he never says why-nor 

why it should be framed in the way he now suggests. In any case, his 

current approach proves on analysis to be as categorical as the more 

sweeping principle he defended in his original text and ultimately no 

more persuasive. 

Begin with its historical implausibility. Like the "broader" argument 

he now disowns, the current variant fails the test of history. In 

the first place, it is in serious tension with-if not entirely irreconcilable 

with-such established and essential practices as agency rulemaking, 

judicial fashioning of federal common law, and the President's 

legislative authority in foreign affairs. If the bicameral congressional 

lawmaking procedure were presumptively exclusive in accordance 

with Professor Tribe's categorical principle, how could Congress delegate 

lawmaking authority to administrative agencies, which in turn 

further discussion, see Henkin, supra note 9, at 44 & 342 n32. Thus, even unenumerated 

powers can be overlapping. 

124 So Professor Tribe indicates in his letter commenting on the manuscript of this Article 

after it was accepted for publication. See Tribe Letter to Golove of Apr. 30,1993, supra 

note 80, at 1-3. 

125 See id. 

126 In his most explicit statement of this narrower claim, Professor Tribe argued: 

[Tihe most plausible way of reading the Constitution as a legal text, in light of 

the historical background against which it was adopted-and particularly in 

light of the overarching concern with state sovereignty that both Article I and 

Article V reflect-would be to read as exclusive those provisions that specify 

how elements of the supreme law of the land are to be adopted. 

Tribe, Taking Text, supra note 3, at 1244. 

M See infra Part HIJ.C.1. 

128 See Tribe Letter to Golove of Apr. 30, 1998, supra note 80, at 1-3 (stating that his 

expressio argument is limited to lawmaking provisions of Constitution, but not mentioning 

federalism concerns in support of this version of argument).



promulgate rules with the binding force of law 129 Likewise, if Article 

I truly provided the exclusive avenue for lawmaking, how could the 

federal courts, in the absence of an act of Congress, fashion rules of 

decision for cases within their jurisdiction At least equally mysterious, 

how could the President, on his own steam, make law in the field 

of foreign affairs 130 Were we to accept the apparent implications of 

Professor Tribe's principle, it is exceedingly difficult to see how these 

practices could survive constitutional scrutiny. 

Nor do the difficulties end here. Perhaps even more starkly, Professor 

Tribe's principle is flatly contradicted by the admiralty power. 

The Constitution expressly grants the courts power to fashion rules in 

this area under the extension of the judicial power "to all Cases of 

131 

admiralty and maritime Jurisdiction.' Although this provision 

specifies a procedure by which supreme law of the land is to be fashioned-and 

although there is no grant of a comparable congressional 

power-the Court in In re Garnett 132 held that Congress's power over 

the subject was "coextensive" with that of the courts, rendering the 

two powers fully interchangeable. 133 And notwithstanding the appar- 

129 See American Trucking Ass'n v. United States, 344 U.S. 298, 310-13 (1953) (upholding 

power of Congress to delegate power to make binding regulations). 

130 For further discussion of these two questions, see supra note 118. 

131 U.S. Const. art. III, § 2, cl. 1. This is the only mention of the admiralty and maritime 

power in the Constitution. 

132 141 U.S. 1 (1891). 

133 Id. at 12; see also Panama R.R. Co. v. Johnson, 264 U.S. 375,385-86 (1924) (reaffirming 

holding in Garnett). The Court in Garnett held that Congress's legislative authority 

over admiralty and maritime law is "coextensive with that law.... [I]n maritime matters, it 

extends to all matters and places to which the maritime law extends." Garnett, 141 U.S. at 

12. 

The admiralty power is a good case in point because of its close parallels to the treaty 

power. Although it has long been an area of plenary federal control, Article I is silent on 

the subject, just as it is on the power to authorize agreements. In contrast, just as Article II 

grants the treaty power to the President and the Senate, Article III expressly grants the 

admiralty power to the federal courts. The Article III grant led the Court in early decisions 

to limit Congress's authority over admiralty to what it could otherwise reach under 

the Commerce Clause. See, e.g., The Daniel Ball, 77 U.S. (10 Wall.) 557, 564-65 (1871) 

(holding that Congress has no power to regulate intrastate shipping over navigable waters); 

Moore v. American Transp. Co., 65 U.S. (24 How.) 1, 39 (1860) (holding that Limitation of 

Liability Act can only apply to vessels engaged in foreign commerce or commerce between 

states); The Passenger Cases, 48 U.S. (7 How.) 283, 400 (1849) (opinion of McLean, J.) 

(stating that Congress could not impose licensing requirements on vessels engaged in intrastate 

commerce). Moreover, the Court initially viewed the judicial role as limited to applying 

the maritime law of nations. See Martin v. Hunter's Lessee, 14 U.S. (1 Wheat.) 304, 

335 (1816) (reasoning that Constitution vests admiralty jurisdiction in federal courts because 

"the law and comity of nations" form an "essential inquiry" in such cases); see also 

American Ins. Co. v. Canter, 26 U.S. (1 Pet.) 511, 545-46 (1828) (Marshall, C.J.) (denying 

that maritime cases arise under Constitution or laws of United States but rather are ruled 

by "the law admiralty and maritime, as it has existed for ages"). As time went on, however, 

the Court took a more assertive role. See, e.g., The Lottawanna, 88 U.S. (21 wall.)




ently irremediable conflict this settled understanding creates with Professor 

Tribe's categorical presumption, he has elsewhere fully affirmed 

this result. 134 

But Professor Tribe's argument fails not only on historical 

grounds. His more "modest" view is also defective on a conceptual 

level. Indeed, like his broader argument, it founders on the shoals of 

the Necessary and Proper Clause. By failing to bear that clause in 

mind, Professor Tribe unjustifiably conflates two separate-and importantly 

different-questions: whether we ought to presume that express 

lawmaking procedures are exclusive of entirely unspecified 

alternative procedures, and whether we ought to indulge the same 

presumption when there are two textually specified procedures that 

558, 576-77 (1874) (asserting expanded conception of Court's role in declaring maritime 

law, while affirming traditional view that only Congress can "make the law"). For an excellent 

historical account of the various metamorphoses in the construction of the admiralty 

and maritime power, see Note, From Judicial Grant to Legislative Power. The Admiralty 

Clause in the Nineteenth Century, 67 Harv. L Rev. 1214, 1230-37 (1954); see also Grant 

Gilmore & Charles L. Black, Jr., The Law of Admiralty 45-47 (2d ed. 1975) (detailing 

sources of U.S. admiralty law). 

Apparently, Gilmore and Black thought the expressio inference linguistically strong, 

referring to the "constitutional puzzle" of plenary congressional power over general maritime 

law and to the Court's reasoning in later cases affirming plenary congressional power 

as "forced." Gilmore & Black, supra, at 47. Of course, this troubled them not an iota. See 

id. (affirming modem practice). Nor has it troubled Professor Tribe. See Tribe, Constitutional 

Law, supra note 11, § 5-3, at 304 & n.13 (same). One might seek to overcome the 

expressio inference by arguing that Congress's necessary and proper power to carry into 

execution the jurisdictional power granted to the courts gives it power to prescribe the law 

to be applied by the courts in maritime cases, although such a view might have problematic 

implications for the Diversity Jurisdiction Clause, see U.S. Const. art. m, § 2, ci. 1. Presumably, 

something like this line of reasoning underlies Professor Tribe's view. But the 

linguistically powerful retort, from the expressio unius side, would be that the Framers, had 

they intended a plenary authority in Congress, would hardly have specified the jurisdictional 

grant but failed to mention the substantive power. In this sense, the expressio imius 

argument against a necessary and proper congressional power over maritime law is 

stronger than the argument against congressional power to authorize agreements. 

Another difference between the admiralty power and the power to approve agreements 

is that the former implies congressional supremacy, whereas the latter does not. 

Indeed, given general limits on the legislative powers of the courts, implied congressional 

power is not only supreme over the explicit judicial power but may even outstrip it in 

extent. Thus, for example, there may well be changes in maritime law that Congress may 

make, but that would be seen as beyond the proper authority of the courts. In contrast, 

like statutes and treaties, congressional-executive agreements and treaties are equal in status. 

As a result, the last in time controls. See The Chinese Exclusion Case, 130 U.S. 581, 

599-600 (1889) (applying later-in-time statute which was inconsistent with earlier treaty 

obligation and declaring that treaties and statutes have equal constitutional status); 

Whitney v. Robertson, 124 U.S. 190, 194 (1888) (declaring that treaties and statutes have 

equal constitutional status and that later in time controls); Restatement, supra note 9, 

§ 115(1), (2) (affirming later-in-time principle); id. § 115 cmt. c (indicating that 

congressional-executive agreements supersede inconsistent prior law or agreement). 

04 See supra note 133.



[Vol. 73:1791 

can plausibly be construed as overlapping in coverage. In the former 

case, it may well be reasonable to begin with a presumption of exclusivity. 

135 But in the latter, context controls, and rigid categorical presumptions 

will simply not avail the interpretive enterprise. In these 

cases, there is no neutral baseline from which to proceed, no way of 

harmonizing the two procedures that we can presume a priori the 

Framers would have preferred. Perhaps, they wished one procedure 

to cover a particular field because they believed special considerations 

required unique procedural protections in that area. On the other 

hand, perhaps they only wished to provide an alternative procedure to 

be utilized in cases where for some reason the usual lawmaking procedure 

proved inconvenient or inadequate to reach the subject matter. 

To be sure, contemporary interpreters may take comfort where the 

Framers made clear whether the procedures they created were to be 

exclusive or concurrent. But where they failed to do so, no categorical 

135 Professor Tribe does not tell us why this should be so, but a number of arguments 

suggest themselves. From an historical perspective, the lawmaking provisions-particularly 

those dealing with statutemaking-were carefully honed to protect a number of 

fiercely competing interests, and the balance struck might be undermined by permitting 

the use of textually unspecified procedures: "Familiar historical materials provide abundant 

support for the conclusion that the power to enact statutes may only 'be exercised in 

accord with a single, finely wrought and exhaustively considered, procedure."' Clinton v. 

City of New York, 118 S. Ct. 2091, 2103-04 (1998) (quoting I.N.S. v. Chadha, 462 U.S. 919, 

951 (1983)). This conclusion is also supported by more theoretical concerns. Consider the 

perspective of the drafters of any well-framed constitution. Arguably, they would be vitally 

concerned to establish the basic rules governing lawmaking. This would be necessary 

to avoid endless contestation over lawmaking procedures and to ensure the realization of 

the procedural protections built into the legislative process. See Christopher I. Eisgruber, 

Should Constitutional Judges Be Philosophers 13 (1997) (unpublished manuscript) (on 

file with the New York University Law Review) (arguing that ideal framers would adopt 

fixed concrete provisions so as to establish basic rules of political game and avoid prolonged 

destructive contestation over procedures). For this reason as well, then, it might be 

reasonable to assume that when a constitution does define those procedures, they ought to 

be deemed exclusive of any textually undefined alternatives. As a corollary to this principle, 

moreover, it is equally sensible to presume that the legislature cannot create alternative 

lawmaking procedures under its general legislative powers, e.g., under the Necessary 

and Proper Clause. Otherwise, a temporary majority could extend or even entrench its 

own power by assigning legislative powers to, say, its preferred political party, or, less dramatically, 

it could undermine the procedural safeguards built into the text-based procedures. 

The Supreme Court seems to agree. See Clinton, 118 S. Ct. at 2102-03, 2107 

(striking down Line Item Veto Act on ground that Article I, Section 7's procedure for 

adopting, amending, and repealing laws is exclusive, thereby ruling out any textually unspecified 

repeal power in President, and noting that "Congress cannot alter the procedures 

set out in Article I, § 7, without amending the Constitution"); cf. A.L.A. Schechter Poultry 

Corp. v. United States, 295 U.S. 495, 521-23, 537 (1935) (striking down congressional delegation 

of legislative authority in National Industrial Recovery Act in part because delegatees 

were private parties, viz., representatives of affected industries). 

Although I cannot pursue the matter here, I believe that the grounds for applying the 

presumption discussed in the text to provisions for constitutional amendments are far 

weaker than the grounds for applying it to basic lawmaking provisions.




presumption can settle the resulting interpretive difficulties. These 

ambiguities can only be resolved by careful consideration of context, 

and even then no construction will necessarily prove most persuasive 

on the basis of text alone. 3 6 

136 The Court's recent decision in Clinton v. City of New York illustrates the distinction 

between these two questions. None of the justices, including the three dissenters, believed 

there was any plausible textual basis for a presidential power to repeal acts of Congress. In 

the absence of any textual basis for the line item veto, they all agreed, unsurprisingly, that 

the Article I, Section 7 procedure for making and repealing laws was exclusive. See 

Clinton, 118 S. Ct. at 2103-04; id. at 2115-16 (Scalia, J., concurring in part and dissenting in 

part); id. at 2121 (Breyer, J., dissenting). The dissenters sought to uphold the Line Item 

Veto Act on entirely different grounds-that it did not authorize the President to repeal 

acts of Congress within the meaning of Article I, but only to exercise delegated legislative 

power, which Congress could properly delegate under its Article I, Section 8 powers. See 

id. at 2116-18 (Scalia, J., concurring in part and dissenting in part); id. at 2125-31 (Breyer, 

I., dissenting). 

Consider, in contrast, the President's legislative powers in foreign affairs. See supra 

note 118. In this area, the President acts not on the basis of a delegation from Congress, 

but on the basis of his own substantive powers over foreign affairs and the implied powers 

doctrine. See infra notes 328-29 and accompanying text for further discussion of the implied 

powers doctrine as applied to the President's powers. Incident to his diplomatic po,;- 

ers, for example, the President has the implied power to determine on a case-by-case basis 

whether the courts should invoke sovereign immunity to dismiss suits against foreign 

states. For further discussion, see supra note 118. Here, then, there is an arguable textual 

basis for overlapping legislative powers in both Congress and the President. As a consequence, 

Professor Tibe's principle of presumptive exclusivity is inapplicable, and the 

Court has found, pace Professor Tribe, that all things considered, the President should have 

concurrent legislative powers in this context. See supra note 118. Ironically, Professor 

Tribe takes no note of the far stronger textual grounds for congressional power to authorize 

the President to approve agreements, given its status as the national legislative body 

and the explicit incorporation of the implied powers doctrine in the Necessary and Proper 

Clause, see supra notes 49-56, 67, and accompanying text; infra notes 235-37 and accompanying 

text, than for the President to make law in the field of foreign affairs. 

For a middle ground example-and one closely related to the line item veto controversy--consider 

the much vexed question of impoundments. Presidents have sometimes 

claimed an executive power, even in the absence of an act of Congress, to decline to spend 

funds appropriated for specific purposes by Congress. Here, the claim is that the 

President's explicit Article H power to execute the law gives him the authority to impound. 

For further discussion, see Tribe, Constitutional Law, supra note 11, § 4-12, at 258 (discussing 

issue and relevant Supreme Court decisions). While I agree that this argument ought 

properly to be rejected, as the Court implicitly did in Train v. City of New York, 420 U.S. 

35, 41-49 (1975) (refusing to construe congressional statute to permit impoundment), this 

position cannot be persuasively defended on the basis of any presumption of exclusivity 

but only on a contextual assessment that textual and structural considerations argue 

strongly against recognizing the power. See Tribe, Constitutional Law, supra note 11, § 4- 

12, at 258-59 (arguing against presidential impoundment power not on basis of any presumption 

of exclusivity but on historical, structural, and textual grounds, including conflict 

impoundment power would create with veto provisions of Article I, Section 7). As 

Professor Tribe has always seemed to recognize in the past, there is simply no way to avoid 

the difficult interpretive questions these kinds of cases present, and categorical presumptions 

will not advance the effort.



[Vol. 73:1791 

Notwithstanding these considerations, Professor Tribe has not 

been careful to notice the distinction between these two questions in 

applying his "narrow" expressio unius principle to the Treaty Clause. 

The question is not whether treaties can be made, say, by a committee 

appointed by the Supreme Court. The Treaty Clause is exclusive in 

that sense-exclusive, that is, of any textually undefined lawmaking 

procedure. What is in issue is whether the Article I lawmaking procedures-procedures 

carefully defined in the text-cover agreementmaking, 

or more precisely, whether they allow Congress to adopt laws 

authorizing the President to conclude international agreements. In 

this context, Professor Tribe's presumptions-as opposed to particularized 

arguments based on the language and structure of the Constitution 

and, I might add, on its interpretive history-are of no help. 

Indeed, this is all the more emphatically the case with the Treaty 

Clause because it defines a specialized procedure for dealing with a 

narrow category of lawmaking and does not specify how it is to interrelate 

with Congress's general legislative powers and its virtually plenary 

authority in the field of foreign affairs. 137 No presumption can 

tell us whether the Framers intended the Treaty Clause to be the exclusive 

means by which international commitments could be tied or 

whether it was only to be an extraordinary procedure affording the 

President an option, to be used in his discretion. 138 Indeed, Professor 

Tribe's retreat to categorical presumptions may best be understood as 

an implicit recognition that the interpretive scales do not tip decidedly 

one way or the other-that is, without the placement of a heavy 

thumb on one side of the balance. Beyond the limited weight which I 

have described, then, application of expressio unius to the Treaty 

Clause adds nothing to the argument for the exclusive view. 

137 For further discussion of Congress's foreign affairs authorities, see supra notes 28, 49- 

52, and accompanying text; infra notes 298-304 and accompanying text. 

138 As Justice Breyer recently put it in reaffirming the pervasiveness of overlapping 

powers: 

This Court has frequently found that the exercise of a particular power, such as 

the power to make rules of broad applicability or to adjudicate claims, can fall 

within the constitutional purview of more than one branch of Government. 

The Court does not "carry out the distinction between legislative and executive 

action with mathematical precision" or "divide the branches into watertight 

compartments," for as others have said, the Constitution "blend[s]" as well as 

"separat[es]" powers in order to create a workable government. 

Clinton, 118 S. Ct at 2123 (1998) (Breyer, J., dissenting) (citations omitted) (quoting 

Springer v. Philippine Islands, 277 U.S. 189, 211 (1928) (Holmes, J., dissenting)); see also 1 

Kenneth Culp Davis, Administrative Law Treatise § 1.09, at 68 (1958) (observing that solutions 

to contemporary administrative challenges may require that executive, legislative, 

and judicial powers be blended in a single agency).




2. An Offense to Good Neighborliness: Ignoring the 

Appointments Clause 

Professor Tribe next argues strenuously that the Appointments 

Clause bears importantly on the proper reading of the Treaty Clause. 

It grants the President the power to appoint principal officers of the 

United States by and with the advice and consent of the Senate, but 

permits Congress to provide separately for the appointments of inferior 

officers. 139 It is common ground, Professor Tribe argues, that the 

advice and consent procedure for confirming the President's nominations 

of principal officers is exclusive. 140 From this he seeks to draw 

two conclusions: First, he thinks that the express nonexclusivity of the 

procedure for confirming inferior officers implies the exclusivity of the 

Treaty Clause because of the latter's failure to include comparable 

language of nonexclusiity. 141 Let's call this the linguistic argument. 

Second, he is emphatic that the fact that the two clauses are "nextdoor 

neighbor[s],"' 142 are really two halves of the same clause, 143 and, 

indeed, are "contained in the same sentence, ' 144 is persuasive grounds 

for giving them a parallel reading. If the procedure for confirming 

principal appointments is exclusive of Congress's Article I, Section 8 

powers, as all allegedly agree, then the procedure for approving treaties 

must be as well. Let's call this the same sentence rule. 

Neither of these arguments is persuasive even on its own terms, 

and they both reflect the highly formalistic character of Professor 

Tribe's methodological approach. Let us begin with the linguistic argument. 

Not only are the inferences he wishes to draw from the Ap- 

139 Specifically, the Appointments Clause provides that the President 

shall nominate, and by and with the Advice and Consent of the Senate, shall 

appoint Ambassadors, other public Ministers and Consuls, Judges of the 

supreme Court, and all other Officers of the United States, whose Appointments 

are not herein otherwise provided for, and which shall be established by 

Law- but the Congress may by Law vest the Appointment of such inferior 

Officers, as they think proper, in the President alone, in the Courts of Law, or 

in the Heads of Departments. 

U.S. Const. art. II, § 2, cL 2. 

140 As we have already seen, however, this claim is not in fact accurate. See supra note 

119. From the outset, presidents have appointed various diplomatic agents qualifying in all 

relevant respects as ambassadors or public ministers, without obtaining the Senate's advice 

and consent. See supra note 119. Of course, the fact that the procedure for appointing 

principal officers is not, as Professor Tribe blithely assumes, wholly exclusive seriously undermines 

his Appointments Clause argument. I nevertheless leave this point aside. Even 

were the Appointments Clause exclusive, as he claims, his argument fails on the various 

other grounds that I elaborate below. 

141 See Tribe, Taking Text, supra note 3, at 1272-73. 

142 Id. at 1272. 

143 See id. at 1273. 

144 Id. at 1274.



[Vol. 73;1791 

pointments Clause weak, they illustrate the manipulability of 

arguments from constitutional silences and the care one must take in 

assessing their probative force. 

Contrary to Professor Tribe's claims, there are crucial differences 

in text and structure between the Appointments and Treaty Clauses 

that render any effort to analogize from one to the other fatuous. 145 

Begin with crucial differences in the language of the two clauses, the 

significance of which Professor Tribe simply dismisses. 146 While the 

first part of the Appointments Clause, much like the Treaty Clause, 

gives the President the power, by and with the advice and consent of 

the Senate, to make appointments of "Officers of the United States, 

whose Appointments are not herein otherwise provided for, and 

which shall be established by Law,"' 1 47 the second part, without parallel 

in the Treaty Clause, adds a proviso specifying a different rule for 

inferior officers: "but the Congress may by Law vest the Appointment 

of such inferior Officers, as they think proper, in the President 

alone, in the Courts of Law, or in the Heads of Departments. 1 48 This 

proviso, as the Supreme Court long ago recognized, strongly suggests 

that Congress has no power over appointments of principal officers 

and that its role in inferior appointments is limited to vesting that 

power in the President, the heads of departments, or the courts. One 

would hardly have expected the Framers to have included the proviso 

had they intended Congress to retain plenary authority under Article 

I, Section 8 over appointments. The language at best would have been 

confusing surplusage. Why explicitly carve out a special category of 

inferior officers for the sole purpose of giving Congress express supervisory 

authority over them when it already has implied supervisory 

authority over appointments generally, including over the very class of 

officers from which the inferior officers category is being shorn The 

proviso would not only be unnecessary; given its obvious implications, 

it would invite misunderstanding, 49 or, as Hamilton put it in a compa- 

145 On the ambiguity of the concept of exclusivity, see supra note 119. 

146 Indeed, he believes that placing any weight on these differences is "particularly odd." 

Tribe, Taking Text, supra note 3, at 1274 n.182. 

147 U.S. Const. art. II, § 2, cl. 2. 

148 Id. 

149 Reading the Appointments Clause as not excluding a congressional role in principal 

appointments would not literally render the proviso surplusage. Other reasons for retaining 

the language can be imagined; they are just implausible for textual and structural reasons. 

For example, it might be suggested that the Framers wanted Congress to retain 

plenary authority over principal appointments but to have a more limited authority over 

inferior appointments. Under this view, the language of the proviso would be limiting, 

specifying a more constricted range of options for Congress over inferior appointments. 

This interpretation, however, stands the Framers' evident intention on its head, turning the 

appointments power into a largely legislative rather than executive function. In any case, it




rable context, "it would be both unnecessary and dangerous." 150 And 

so the Supreme Court recognized in Myers v. United States, 151 the 

leading case construing the Appointments Clause: 

Here, then, is an express provision [the proviso], introduced in 

words of exception, for the exercise by Congress of legislative 

power in the matter of appointments and removals in the case of 

inferior executive officers. The phrase "But Congress may by law 

vest" is equivalent to "excepting that Congress may by law vest." 

By the plainest implication it excludes Congressional dealing with 

appointments or removals of executive officers not falling within the 

exception .... 

... Article II expressly and by implication withholds from Congress 

power to determine who shall appoint and who shall remove 

except as to inferior offices.' 5 2 

founders on the shoals of the text itself. Finding a limitation rather than an augmentation 

in a clause that begins "but the Congress may.. .," id. art. II, § 2, cL 2, is not just linguistically 

awkward, but virtually textually absurd. The structure of the clause, moreover, is 

inconsistent with this reading. It authorizes-"but the Congress may. ."----but does not 

require Congress to treat some officers as inferior. Congress, however, would have no 

reason ever to do so if by deeming an officer inferior it thereby limited its range of options. 

In other words, if the proviso was a limitation on, rather than an expansion of, Congress's 

powers, then the Framers would certainly not have left the invocation of those limits entirely 

to Congress's discretion. Alternatively, it might be argued that the Framers assumed 

that under Article I Congress would have the power to vest appointments in itself, either 

by reserving to itself the power to appoint or by requiring its approval of the President's 

nominations, but that it would not have power to delegate the appointments power to any 

other organ of government- The proviso, then, would expand Congress's power over appointments 

of inferior officers by giving it vesting options that would not apply to principal 

officers. This argument, however, simply trades one implausibility (that the proviso constricts 

rather than expands congressional power) for another:. If the Framers bad intended 

to give Congress Article I legislative power to make appointments, Congress would have as 

much authority to delegate this power to, say, the President or the heads of departments, 

as to retain the power in itself. Its Article I powers, then, would engulf the proviso, rendering 

it surplusage. As above, of course, this reading of the text would also stand the principle 

of the separation of powers on its head. 

150 The Federalist No. 32, supra note 94, at 199 (referring to clause in Article I, Section 

10 barring states without consent of Congress from laying any imposts or duties specifically 

on imports or exports, and arguing that it would not have been included had affirnative 

Article I, Section 8 power of Congress to lay and collect taxes, duties, and imposts implicitly 

prohibited states from imposing taxes on any articles whatsoever). This answers Professor 

Tribe's concern about Congress vesting the appointment of a Trade and Commerce 

Secretary, a principal officer, in the Supreme Court. See Tribe, Taking Text, supra note 3, 

at 1274-75. Such a power is excluded because of the clear implications of the proviso. The 

Framers would not have expressly granted Congress the power to vest the appointment of 

inferior officers in, inter al/a, the courts, if Congress had in any case retained general Article 

I authority to vest appointments of both principal and inferior officers in the courts. 

For further discussion, see infra note 319. 

151 272 U.S. 52 (1926). 

152 Id. at 127, 129.



[Vol. 73:1791 

As to inferior officers, moreover, there are equally compelling textual 

reasons to limit Congress's supervisory powers strictly to the three 

vesting options specified in the proviso. To concede Congress any 

broader role would render the specific grants in the proviso surplusage. 

Why specify Congress's authority to vest inferior appointments 

in these three ways if Congress retained plenary authority to vest 

them anywhere it pleased 153 

In light of these textual considerations, it is hard to understand 

why Professor Tribe believes the language of the Appointments 

Clause lends weight to the argument for an exclusive reading of the 

Treaty Clause. The peculiar textual features that lead to an exclusive 

reading of the former clause do not apply to the latter. It contains no 

"proviso" or other comparable language which would have clarified 

the Framers' intent, no express conferral upon Congress of authority 

over some types of agreements which would be surplusage were Congress 

to retain Article I, Section 8 authority over all other agreements. 

The text simply grants the President the power, by and with the advice 

and consent of the Senate, to make treaties. The contrast with the 

Appointments Clause could not be more striking. 

In an effort to avoid the obvious implications of the proviso, Professor 

Tribe resorts to high formalism. Reversing the intuitive inference, 

he argues that it is the proviso itself which supports the 

exclusivity of the Treaty Clause, because it demonstrates that the 

Framers would have been explicit had they intended the treaty power 

to be nonexclusive. The argument proceeds in two steps. First, rather 

than considering how the provisions for the appointment of principal 

and inferior officers are substantively intertwined, he treats them as 

two distinct powers and focuses on their formal linguistic properties.' 

54 Thus, he ignores the central point that a nonexclusive reading 

of the procedure for confirming principal officers would render the 

proviso substantively superfluous, instead emphasizing that the proviso 

explicitly authorizes alternative means of appointing inferior officers. 

He then extrapolates from this express nonexclusivity as to 

inferior officers to the conclusion that the Framers would have been 

explicit about the nonexclusivity of the principal officers procedure 

153 See Freytag v. Commissioner, 501 U.S. 868, 883-84 (1991) (holding that Congress 

may only vest appointments of inferior officers in three ways specified in Appointments 

Clause); cf. Buckley v. Valeo, 424 U.S. 1, 127 (1976) (noting that "[w]hile the Clause expressly 

authorizes Congress to vest the appointment of certain officers in the 'Courts of 

Law,' the absence of similar language to include Congress must mean that neither Congress 

nor its officers were included within the language 'Heads of Departments' in this part 

of cI. 2"). 

154 See Tribe, Taking Text, supra note 3, at 1272-75.




had they so intended. 5 5 Taking it one step further, he then extrapolates 

to the Treaty Clause: Its failure to be expressly nonexclusive 

likewise argues in favor of its exclusivity. 15 6 Thus, for Professor Tribe, 

the critical textual feature is the use in one instance of language of 

nonexclusivity and the failure to use the same or comparable language 

in others. 

This focus on form over substance is a notable characteristic of 

Professor Tribe's argumentation throughout. While a staple of the 

lawyerly trade, however, it is at best exceedingly incomplete as an interpretive 

methodology for constitutional law. "It is," after all, "a 

constitution we are expounding."' 5 7 By presuming consistency on the 

highest levels of linguistic abstraction, such rigid interpretive rules demand 

more of the text than it can possibly offer. As a result, Professor 

Tribe's approach also renders the text hopelessly self-defeating: 

From the Constitution's express language of exclusivity in its grant to 

the Senate of the "sole Power to try all Impeachments, 

158 

for example, 

can we argue persuasively in favor of the nonexclusivity of other 

constitutional grants of power that lack its explicitness If not all 

other grants, what about grants of power, like the Treaty Clause, 

which are given to the Senate Professor Tribe does not tell us why 

this implication is any less textually persuasive than his effort to draw 

an exclusive reading of the Treaty Clause from the multiple procedures 

provided for making inferior appointments. 159 More impor- 

155 See id. at 1273 (noting that "Clause 2's affirmative authorization for Congress to 

alter the procedures for appointing inferior officers suggests that the Constitution would be 

explicit if the prescribed methods of confirming principal officers were not exclusive"). 

156 See id. at 1272 (arguing that "there is a telling difference between the Treaty Clause 

and the immediately adjacent Appointments Clause: only the Appointments Clause provides 

for alternative consent procedures"). 

157 McCulloch v. Maryland, 17 U.S. (4 Wheat.) 316, 407 (1819) (Marshall, CJ.). 

158 U.S. Const. art. I, § 3, cl. 6 (emphasis added). Consider as well Article I, Section 2, 

Clause 5 which declares that the House "shall have the sole Power of Impeachment." Id. 

art. I, § 2, cl. 5 (emphasis added). 

159 Similarly, the Exclusive Legislation Clause, id. art. I, § 8, cl. 17, empowers Congress 

to "exercise exclusive Legislation" over the district ceded by states for the seat of the government 

and likewise over places purchased from states for the purpose of erecting forts, 

government buildings, and the like. Id. (emphasis added). Presumably, this was meant to 

emphasize that the states which ceded the district or sold the places would have no 

continuing legislative authority over them. See generally Tribe, Constitutional Law, supra 

note 11, § 5-11, at 327-29 (discussing interpretation of Exclusive Legislation Clause). 

Would Professor Tribe conclude, then, that all of the other grants in Article I, Section 8 are 

nonexclusive, permitting concurrent state regulation of their subject matter in the absence 

of congressional preemption After all, none contain similarly explicit language of exclusivity. 

For further discussion, see infra note 209. 

Ironically, despite the explicit language of exclusivity in the Exclusive Legislation 

Clause, the Supreme Court has permitted states to exercise concurrent legislative authority 

in certain respects. See, e.g., United States v. State Tax Comm'n, 412 U.S. 363, 369-71



[Vol. 73:1791 

tantly, it is common ground, I assume, that some constitutional grants 

of power are exclusive while others are not. 160 In most cases, however, 

there is no explicit language one way or the other, and we are 

left to construe the sounds of silence. Ordinarily, it will not advance 

the effort to invoke the presence or absence of linguistic formulas 

taken from the few instances in which the Framers adopted explicit 

language of exclusivity or nonexclusivity. If we are to take our interpretive 

responsibilities seriously, we will have to consider each provision 

in context and examine its logical interrelationships with other 

possibly overlapping provisions. 161 

(1973) (holding that previously enacted state law continues to govern state property acquired 

by federal government with state's consent unless inconsistent with federal policy or 

preempted by Congress); Pacific Coast Dairy, Inc. v. Department of Agric., 318 U.S. 285, 

294 (1943) (same); Tribe, Constitutional Law, supra note 11, § 5-11, at 328-29 (discussing 

cases). These holdings only suggest again how far Professor Tribe's current approach is 

out-of-step with our tradition of constitutional interpretation. 

It is notable as well that Article III, Section 3, Clause 1 provides that treason against 

the United States "shall consist only in levying War against them, or in adhering to their 

Enemies, giving them Aid and Comfort." U.S. Const. art. III, § 3, cl. 1 (emphasis added). 

Does this mean that when the Framers wished to make an exclusive list they used the word 

"only" Professor Tribe's approach certainly makes constitutional interpretation simpler. 

Yet, it is hard to believe that even he would defend the results. See infra note 403 for 

further discussion of Professor Tribe's concerns about missing "only[s]." 

160 See supra notes 117-23, 131-34, and accompanying text (discussing numerous examples 

of both exclusive and overlapping powers); infra notes 194-209,398-403, and accompanying 

text (same). 

161 There is an additional irony here. As we shall see, Professor Tribe goes to some 

lengths to defend the unilateral executive agreement even as he attacks the congressionalexecutive 

agreement. He claims that the President may make less important agreements 

on his own authority but that important agreements must receive senatorial advice and 

consent. See infra Part III.D.2. If a linguistic analysis of the Appointments Clause teaches 

us anything about the Treaty Clause, however, it is that we ought to be skeptical about the 

President's claimed unilateral agreement-making powers-and likewise about any effort to 

restrict Congress to approving only less important agreements. Recall that the Teaty 

Clause says nothing about the power to conclude undertakings other than through the 

Senate supermajority method, whereas the Appointments Clause carefully distinguishes 

between principal (read, "important") and inferior (read, "less important") officers and 

explicitly allows inferior officers to be appointed through alternative less burdensome procedures, 

viz., by the President alone. The absence of a comparable distinction in the Teaty 

Clause argues against finding that the President has the power to conclude agreements on 

his own authority, so long as they are, in effect, "inferior" rather than "principal" agreements, 

and it likewise argues against limiting Congress to approving such "inferior" agreements. 

Under Professor Tribe's approach, the proviso to the Appointments Clause 

strongly suggests that the Framers would have been express about authorizing the 

President to make inferior agreements on his own. Likewise, they would have been express 

had they wished either, depending upon one's view of congressional Article I power, 

to limit Congress to approving only inferior agreements or, conversely, to authorize it to 

approve only inferior agreements. 

Despite these textual implications, however, when the WTO Agreement was pending 

before Congress, Professor Tribe, even while relying upon the Appointments Clause, made 

precisely the argument that it seems to rule out-that the President and the Congress may



AGANST FREE-FORM FORMALISM 

Of course, courts never rely solely upon purely textual considerations 

in resolving great constitutional questions, not even the more 

nuanced inquiry that looks to context and substance rather than simply 

linguistic forms. The Appointments Clause is no exception. It is 

no small irony that Professor Tribe begins his article on the Treaty 

Clause by elaborating a multidimensional interpretive model that emphatically 

stresses the importance not only of text but of structure and 

architecture. 162 Yet, inexplicably, in his discussion of the Appointments 

Clause, he ignores the pressing structural and architectural considerations 

that give substance to the Court's decisions and which 

sharply distinguish the appointments and treaty powers. 1 63 Chief 

Justice Taft in Myers v. United States,164 with firm support in historical 

make less important agreements but not important treaties. See GATr Hearings, supra 

note 41, at 301 (suggesting that only most important agreements need be submitted to 

Senate rather than Congress); Tribe Letter to Sen. Byrd of July 19,1994, supra note 41, at 2 

(same); Tribe Memo of Oct. 5, 1994, supra note 41, at 8 (same). In response, Professor 

Ackerman and I pointed out the textual inconsistency. See Ackerman & Golove, supra 

note 5, at 923 n.517. Although Professor Tribe now derides our argument as "downright 

silly," Tribe, Taking Text, supra note 3, at 1274 n.182, he has in the meantime changed his 

view about congressional power to meet our point: Professor Tribe now concedes that 

Congress can either approve all agreements falling within its substantive powers or none; it 

is the middle ground that is untenable. He has simply chosen none where we chose interchangeability. 

With respect to unilateral presidential agreements, moreover, he charges us 

with silliness only because he persists in thinking that the Treaty Clause implicitly incorporates 

a distinction between "treaties" and "agreements," and that the clause applies only to 

the former. The President, he says, cannot make unimportant "treaties," only "agreements," 

although the latter are just unimportant treaties. As we will see, this premise is 

entirely untenable, see infra Part M.D.2, and is even contradicted by Professor Tribe himself, 

see infra note 366 and accompanying text. 

Professor Tribe's unwillingness to accept the obvious linguistic implications of the 

Appointments Clause from the point of view of his own interpretive methodology is, of 

course, a consequence of the imperative to uphold the unilateral executive agreement. 

That imperative teaches another lesson that he seems to miss: We ought to be skeptical 

about the strength of any inferences drawn on the basis of the kinds of purely linguistic 

considerations that he purports to find definitive. At this level of abstraction, the text 

simply refuses to cooperate. In my view, whether the President has unilateral powers to 

make some kinds of agreements cannot be intelligently resolved by a close parsing of the 

Appointments Clause. Nor-and this is the crucial point--can Congress's powers to approve 

international agreements. 


163 As Justice Scalia put it "[T]he basic separation-of-powers principles ... are what 

give life and content to our jurisprudence concerning the President's power to appoint and 

remove officers." Morrison v. Olsen, 487 U.S. 654,715 (1938) (Scalia, J., dissenting). It is 

not the case, of course, that Professor Tribe is unaware of these structural and architectural 

considerations. He has insightfully discussed them on numerous past occasions. See, e.g., 

Tribe, Constitutional Law, supra note 11, at 213-18,244-54 (discussing separation of powers 

concerns underlying appointment and removal powers). 

164 272 U.S. 52 (1926).



[Vol. 73:1791 

understandings, 165 recognized appointments as a core executive function.166 

Indeed, it was the Convention's decision to include the Senate-inserting 

a branch of the legislature into an executive 

procedure-that stood in need of justification. As Madison said during 

the famous House debate in 1789 over the removal power, this 

legislative encroachment on the executive sphere had to be narrowly 

construed. 67 At stake, of course, is the ability of the executive to 

control the character of the officers who are to carry out his constitutional 

responsibility to faithfully execute the laws. The need to assure 

the independence of the executive against legislative aggrandizement 

and the need to avoid a concentration of power that might threaten 

the liberties of the people were, famously, central preoccupations of 

the Framers. 68 In any case, Madison's principle of strict construction 

165 See, e.g., 1 Annals of Cong. 496-97, 581-82 (Joseph Gales ed., 1789) [hereinafter 

Annals] (remarks of Rep. Madison) (indicating that appointments and removals are executive 

functions); 1 id. at 557 (remarks of Rep. Baldwin) (same); 1 id. at 561 (remarks of Rep. 

Sylvester) (same); see also Myers, 272 U.S. at 115-32 (reviewing removal debate in First 

Congress); id. at 136-57 (reviewing subsequent practice and judicial decisions). 

166 See Myers, 272 U.S. at 115-29 (holding that removal is an executive function). 

167 As Madison put it: 

Perhaps there was no argument urged with more success, or more plausibly 

grounded against the Constitution, under which we are now deliberating, than 

that founded on the mingling of the Executive and Legislative branches of the 

Government in one body. It has been objected, that the Senate have too much 

of the Executive power even, by having control over the President in the appointment 

to office. Now shall we extend this connexion between the Legislative 

and Executive departments, which will strengthen the objection, and 

diminish the responsibility we have in the head of the Executive 

... [The legislature] ought to have nothing to do with designating the man to 

fill the office. That I conceive to be of an Executive nature. Although it be 

qualified in the Constitution, I would not extend or strain that qualification 

beyond the limits precisely fixed for it. We ought always to consider the Constitution 

with an eye to the principles upon which it was founded. 

1 Annals, supra note 165, at 380, 582 (remarks of Rep. Madison); see also Myers, 272 U.S. 

at 120-21, 128-29 (endorsing Madison's argument); 1 Annals, supra note 165, at 557 (remarks 

of Rep. Baldwin) (concurring with Madison). 

168 These separation of powers concerns have underwritten the Court's several decisions 

in this area, which develop the separation of powers theme at great length. See Weiss v. 

United States, 510 U.S. 163, 183-89 (1994) (Souter, J., concurring) (upholding procedure 

for appointing military judges); Freytag v. Commissioner, 501 U.S. 868, 882-92 (1991) (upholding 

procedure for appointing special Tax Court trial judges); id. at 901-08 (Scalia, J., 

concurring); Morrison, 487 U.S. at 685-96 (upholding independent counsel law); Bowsher 

v. Synar, 478 U.S. 714,721-27 (1986) (striking down certain provisions of Gramm-Rudman- 

Hollings Act on ground that congressional control over removal of Comptroller General 

violated separation of powers); Buckley v. Valeo, 424 U.S. 1, 120-29 (1976) (per curiam) 

(striking down Federal Election Campaign Act procedure for appointing commissioners to 

Federal Election Commission); Myers, 272 U.S. at 116-32 (striking down provision that 

President could remove certain postmasters only upon senatorial consent); see also 'Tibe, 

Constitutional Law, supra note 11, §§ 4-9 to 4-10 (affirming centrality of separation of




led the Myers Court to rule out altogether the participation of the 

Senate in removing officers, despite the absence of any explicit textual 

guidance and despite the rule, accepted by the Court, that the power 

to remove is incident to the power to appoint. 169 Indeed, even as to 

inferior officers, over whom Congress has express supervisory authority, 

the Court held that Congress could not 

draw to itself, or to either branch of it, the power to remove or the 

right to participate in the exercise of that power. To do this would 

be to go beyond the words and implications of that clause, and to 

infringe the constitutional principle of the separation of governmental 

powers. 170 

It should be obvious that application of this rule of strict construction 

virtually compels the conclusion that the Appointments Clause is exclusive: 

Notwithstanding Congress's powers under Article I, the legislative 

branch has no role in appointments beyond that explicitly 

specified in the language of the proviso, narrowly construed. 171 

powers concerns in this area). Even in his article, Professor Tribe notices the separation of 

powers underpinnings to the Appointments Clause but only when he is not actually discussing 

his Appointments Clause argument. See Tibe, Taking Text, supra note 3, at 1238 

(noting separation of powers concerns embodied in Appointments Clause in context of 

discussion of legislative veto as example of distortion of constitutional topology). 

169 See Myers, 272 U.S. at 119-22, 163-64. Indeed, the Court was willing to abide congressional 

supervision only over removals of inferior officers whose appointments Congress, 

in accordance with the proviso, had vested outside the normal senatorial procedure. 

Here, the power to vest the appointment yielded the power to control removals as well. 

See id. at 160-61. Even within this narrow category, however, the Court doubted strongly 

whether Congress could exercise removal power over appointments it had vested in the 

President alone. Instead, Congress's power applied only to appointments the proviso permitted 

Congress to vest in the heads of departments, and, the Court insisted, the power 

applied only in cases where Congress had in fact vested the appointment in the head of a 

department. It was not enough that Congress could have vested the appointment but had 

chosen not to do so. See id. at 161-62, 164. Even in these cases, moreover, the Court 

implied what later cases confirmed-that Congress itself can play no direct role in removals. 

While in these cases it may vest the removal power elsewhere than in the President 

and restrict its exercise to specified causes, Congress may not itself participate in the removal 

process. See id. at 161; see also Morrison, 487 U.S. at 685-86 (noting that Congress 

has historically been denied the right to remove officers directly); Bowsher, 478 U.S. at 

722-26 (declaring that "[a] direct congressional role in the removal of officers charged with 

the execution of the laws .. is inconsistent with separation of powers"). 

170 Myers, 272 U.S. at 161; see also Morrison, 487 U.S. at 685-86 (affirming same); 

Bowsher, 478 U.S. at 722-26 (same); Buckley, 424 U.S. at 129, 134-36 (same). 

171 Note that subsequent decisions have retreated somewhat from the Myers highwater 

mark of exclusive executive powers. In Humphrey's Executor v. United States, 295 U.S. 

602 (1935), the Court disapproved Myers in this respect and recognized that Congress 

could restrict the grounds upon which the President can remove even certain principal 

officers, provided their functions are quasilegislative or quasijudicial, rather than purely 

executive, in character. See id. at 629-31. More recently, the Court has abandoned the 

effort to draw bright line distinctions and has permitted Congress to impose limits on removals 

so long as these do not interfere with the President's ability to accomplish his executive 

responsibilities. See Morrison, 487 U.S. at 690-91. Implicitly, then, the Court has



Indeed, wholly independently of textual considerations, these 

structural concerns have proven powerful enough to justify a quarantine 

against any legislative involvement in the appointments process. 

The need to resolve this question of fundamental principle was provoked 

by the constituent instrument Congress promulgated for the 

Philippine Islands during the period of our colonial rule. 172 Unsurprisingly, 

it looked quite a bit like our own Constitution, but for our 

purposes it differed in one critical respect: The Governor General had 

no general power to appoint officers by and with the consent of the 

Philippine Senate but only when the legislature so decided. 173 Rushing 

to test the limits, the legislature appointed the President of its Senate 

and the Speaker of its House, along with the Governor General, to 

recognized that Congress has a reserve of Article I authority over removals that permits it 

some supervisory role where the separation of powers does not compel otherwise. 

On the other hand, in deference to the core separation of powers issues at stake, the 

Court has continued to insist that Congress itself never play any direct role in removals. 

See id. at 685-86; Bowsher, 478 U.S. at 722-26. The Court's concerns are not limited to 

ensuring that the President has sufficient control over executive officers to carry out his 

executive duties effectively. In accordance with the Madisonian theory of separated powers, 

it has also worried about the dangers posed by congressional power over the execution 

of the laws. Much like the legislative veto struck down in I.N.S. v. Chadha, 462 U.S. 919 

(1983), Congress's holding of the removal power, the Court has feared, would give it effective 

control over executive officers and thereby over the execution of the laws. See 

Bowsher, 478 U.S. at 726. Such control would enable Congress to usurp the executive 

function and would pose the dangers of unitary, parliamentary government the Framers so 

carefully sought to avoid. See Tribe, Constitutional Law, supra note 11, § 4-10, at 253-54; 

Tribe, Choices, supra note 39, at 74-76. No doubt, these concerns, combined with the textual 

points noted above, would continue to rule out any congressional role over appointments 

beyond that specified in the proviso. 

172 See Springer v. Philippine Islands, 277 U.S. 189 (1928) (construing Phillipine Organic 

Act). 

173 Section 21 of the Organic Act provided: "He shall, unless otherwise herein provided, 

appoint, by and with the consent of the Philippine Senate, such officers as may now be 

appointed by the Governor General, or such as he is authorized by this Act to appoint, or 

whom he may hereafter be authorized by law to appoint." Act of Aug. 29, 1916, ch. 416, 

§ 21, 39 Stat. 545, 552 (emphasis added), made obsolete by Proclamation No. 2695, 3 

C.F.R. 64 (Supp. 1946) (recognizing Philippine independence), reprinted in 22 U.S.C. 

§ 1394 (1994). This provision contrasts with the language of Article II, Section 2 of the 

Constitution, which provides that the President shall appoint, by and with the advice and 

consent of the Senate, "all other Officers of the United States, whose Appointments are 

not herein otherwise provided for, and which shall be established by Law." U.S. Const. 

art. II, § 2, cl. 2. Thus, under the U.S. Constitution, Congress's power is clearly limited to 

creating the office, at which point the President's appointment power automatically kicks 

in, whereas in the Organic Act, the legislature seemed to have the power both to create the 

office and to decide whether to vest the appointment in the Governor General. Note further 

that the Organic Act also vested the supreme executive power in the Governor General, 

granted him general supervision and control over all departments and bureaus of the 

Philippine government, charged him with faithfully executing the laws, and provided that 

all executive functions must be directly under the Governor General or within one of the 

executive departments under his supervision and control. See Springer, 277 U.S. at 200-01.


December 1998] AGAINST FREE-FORM FORMIALISM 

1847 

vote the shares of certain publicly held corporations. 174 This legislative 

usurpation of a core executive function, the United States 

Supreme Court held in Springer v. Philippine Islands, 175 violated a basic 

principle of the separation of powers: "Legislative power, as distinguished 

from executive power, is the authority to make laws, but 

not to enforce them or appoint the agents charged with the duty of 

such enforcement. The latter are executive functions. 

'176 

This was 

sufficient to condemn the legislative appointments even apart from 

"whether the duties devolved upon these members are vested by the 

Organic Act in the Governor-General." 177 Since Springer, the Court 

has consistently enforced this basic separation of powers principle, refusing 

to permit Congress to insert itself into the appointments or removal 

processes, whether by vesting an appointment or removal in 

itself or in one of its officers or by requiring its approval of a presidential 

appointment or removal. 78 

These fundamental separation of powers considerations, which 

underwrite the Court's narrow construction of congressional power 

over appointments, are entirely inapt when applied to the treaty 

power. There is persuasive evidence that a number of key Framers 

viewed the power to approve treaties as properly a legislative, not an 

executive function, and there is strong textual support for this view. 179 

In contrast to the appointments power, international agreementmaking 

was not conceived of as a core executive function. Consequently, 

no emanation from the Treaty Clause can underwrite a 

Myers-like strict rule of construction, ruling out a priori any congressional 

participation in agreement-making as an infringement on the 

174 See Springer, 277 U.S. at 197-99. 

175 277 U.S. 189 (1928). 

176 Id. at 202. 

177 Id. The Court nevertheless went on to hold that the Organic Act, particularly the 

section requiring all executive functions to be under the Governor General or vithin one 

of the executive departments under his supervision and control, prohibited legislative appointments. 

In the process, the Court expressly rejected the application of expressio unitls 

to the listing of the Governor General's appointment powers in Section 21-despite the 

fact that its interpretation seemed to render that listing mere surplusage and thus 

presented a particularly compelling linguistic case for application of the canon. See id. at 

206. 

178 See Bowsher, 478 U.S. at 722-26 (striking down certain provisions of Gramm- 

Rudman-Hollings Act because of Congress's retention of power to remove Comptroller 

General, who was responsible for executing Act); Buckley v. Valeo, 424 U.S. 1.126-28,134- 

37 (1976) (per curiam) (striking down act creating Federal Election Commission because 

Congress vested appointment of some commissioners in its own officers and subjected all 

appointments to congressional approval); see also Morrison, 487 U.S. at 6S5-86 (affirming 

restrictions on congressional involvement). 

179 See infra notes 244-58 and accompanying text.



[Vol. 73:1791 

essential powers of the executive.' 80 As Madison put it, in rejecting 

180 This does not mean that the two clauses ought invariably to be interpreted in parallel 

but opposite fashion. Formal symmetries and oppositions rarely recommend themselves in 

constitutional interpretation. Consider the cognate issues of the removal of executive officers 

and the termination of treaties. Myers and its progeny, as we have seen, strictly limit 

congressional participation in removals of executive officers. See supra notes 169-71 and 

accompanying text. Somewhat similarly, although the issue is still unsettled, see Goldwater 

v. Carter, 444 U.S. 996, 996 (1979) (Powell, J., concurring) (dismissing on justiciability 

grounds challenge to a unilateral presidential treaty termination); id. at 1002 (Rehnquist, 

J., concurring) (same), the leading though still highly controversial modem view is that the 

power to terminate treaties belongs solely to the President. See Restatement, supra note 9, 

§ 339 (affirming President's unilateral power); id. § 339 cmt. a & reporters' note 1 (elaborating 

on same); Henkin, supra note 9, at 211-14 (same); Louis Henkin, Litigating the 

President's Power to Terminate Treaties, 73 Am. J. Int'l L. 647, 651-54 (1979) (same); see 

also Goldwater v. Carter, 617 F.2d 697 (D.C. Cir.) (en banc) (per curiam) (ruling in favor 

of President on narrow grounds in hodgepodge opinion rehearsing wide range of pro- 

Executive arguments), vacated, 444 U.S. 996 (1979) (mem.). But see Goldwater, 444 U.S. 

at 1004 n.1 (Rehnquist, J., concurring) (noting various procedures, including congressional 

authorization or direction, that have been used to terminate treaties); Van der Weyde v. 

Ocean Transp. Co., 297 U.S. 114, 117-18 (1936) (upholding termination pursuant to congressional 

direction and appearing to approve practice). Presidents and their boosters, 

unsurprisingly, have from time to time been unable to resist the temptation to cite Myers in 

support of unilateral presidential authority. See, e.g., Points and Authorities in Support of 

Defendants' Motion to Dismiss or, in the Alternative, for Summary Judgment, Goldwater 

v. Carter, 617 F.2d 697 (D.C. Cir. 1979) (No. 78-2412), reprinted in 2 United States Foreign 

Relations Law 532, 551-52 (Michael J. Glennon & Thomas M. Franck eds., 1980). However, 

the more authoritative commentators reach this conclusion concededly not on the 

basis of the text's original meaning but on "the nature of his office as it has become," 

Henkin, supra, at 652, and "as it has developed over almost two centuries," Restatement, 

supra note 9, § 339 reporters' note 1 (citing, inter alia, United States v. Curtiss-Wright 

Export Corp., 299 U.S. 304, 320 (1936)). 

Indeed, the original understandings, which far better reflect the original meaning of 

the text, were quite to the contrary and strongly confirm the vast gulf between the appointments 

and treaty powers. The decision of 1789, as Myers referred to the famous debate in 

the First Congress, confirmed the view that appointments and removals are core executive 

functions in which Congress may not participate. See Myers v. United States, 272 U.S. 52, 

163 (1926). In sharp contrast, Congress quickly claimed for itself the power to terminate 

treaties when, in 1798, it unilaterally abrogated our alliance with France formalized in the 

Treaties of 1778. See Act of July 7, 1798, ch. 67, 1 Stat. 578 (declaring "the United States 

are of right freed and exonerated from the stipulations of the treaties, and of the consular 

convention, heretofore concluded between the United States and France"). While that 

precedent has, strictly speaking, never been repeated, the predominant practice until at 

least President Wilson and perhaps until the New Deal was either for Congress, or less 

commonly for two-thirds of the Senate, to authorize or direct the President to terminate 

treaties. For historical accounts, compare Goldwater, 617 F.2d at 716,723-32 (MacKinnon, 

J., dissenting in part, concurring in part) (concluding that "Congressional participation in 

termination has been the overwhelming historical practice"), vacated, 444 U.S. 996 (1979) 

(mem.), with David Gray Adler, The Constitution and the Termination of Treaties 149-90 

(1986) (concluding there has been "no predominant method of termination, or even a discernible 

trend," and that there is "no support" for contention that unilateral executive 

termination has dominated since 1920s), and Memorandum from Herbert J. Hansel, Legal 

Adviser, Dep't of State, to Secretary of State (Dec. 15, 1978) (concluding that while early 

precedent is mixed, practice since 1920 confirms President's unilateral power to terminate 

treaties), reprinted in 2 United States Foreign Relations Law, supra, at 377, 382-404. For




1849 

the claim that the strict rule of construction he had argued for in the 

appointments context in the first Congress applied as well to legislative 

involvement in the treaty and war powers: 

To justify any favourable inference from [the removal] case, it must 

be shown, that the powers of war and treaties are of a kindred nature 

to the power of removal, or at least are equally within a grant 

of executive power. Nothing of this sort has been attempted, nor 

probably will be attempted. Nothing can in truth be clearer, than 

that no analogy, or shade of analogy, can be traced between a 

contrasting views of leading historical figures, compare Jefferson, supra note 47, § 52, at 

300 (asserting that "an act of the legislature alone can declare [treaties] infringed and rescinded"), 

and Ware v. Hylton, 3 U.S. (3 Dall.) 199,256,261 (1796) (Iredell, J.) (reprinting 

Justice Iredell's opinion for lower court) (stating that Congress alone has authority to terminate 

treaties), with The Federalist No. 64, at 394 (John Jay) (Clinton Rossiter ed., 1961) 

(indicating that "they [the President and the Senate] who make treaties may alter or cancel 

them"), and Techt v. Hughes, 128 N.E. 185,192 (N.Y. 1920) (Cardozo, J.) (explaining that 

"President and Senate may denounce treaty, and thus terminate its life"), and The Amiable 

Isabella, 19 U.S. (6 Wheat.) 1, 75 (1821) (Story, J.) (stating that "the obligations of the 

treaty could not be changed or varied, but by the same formalities with which they vere 

introduced, or, at least, by some act of as high an import, and of as unequivocal an authority," 

thus arguably suggesting that Congress, as well as President and Senate, may terminate 

treaties). Corwin's early view, which reflected the practice up to that time, placed the 

power in the Congress alone. See Corwin, President's Control, supra note 119, at 111-16 

(reasoning that this exclusive congressional prerogative "flows naturally, if not inevitably, 

from the power of Congress over treaty provisions in their quality as 'law of the land'"). 

The text itself is radically indeterminate on the subject. See generally Adler, supra, at 89- 

111 (canvassing views of Framers and Supreme Court precedents on location of treaty 

termination power). 

In any case, even the modem view favoring unilateral presidential authority is far 

more tolerant of congressional supervision of treaty terminations than the Court has ever 

been of a comparable congressional role in removals. Thus, for example, contrary to the 

clear implication of Myers with regard to appointments, both the Restatement and Professor 

Henkin recognize the power of the Senate to consent to a treaty on the condition that 

the President not terminate it without the approval of either the Senate or Congress. See 

Restatement, supra note 9, § 339 cmt. a & reporters' note 3; Henkin, supra note 9, at 2124 

see also Tribe, Taking Text, supra note 3, at 1253 n.108 (noting that "[s]erious questions 

might be raised by such unilateral termination if a treaty provided for termination exclusively 

by other means"). Furthermore, both likewise accept that the President may not 

exercise a unilateral power to terminate a treaty when doing so would create a serious 

danger of war "in view of the authority of Congress to decide for war or peace." Restatement, 

supra note 9, § 339 cmt. a; see also id. § 339 reporters' note 1 (similar). This latter 

exception, of course, provides fertile grounds for growth. Why not apply the same logic to 

the termination of a treaty that would seriously affect congressional policy on foreign commerce 

or any other matter within its plenary foreign affairs powers 

Finally, the functional differences between the appointments and treaty powers also 

account for the different rules concerning "reservations." The Senate's right to condition 

its consent to treaties on reservations is virtually unlimited. In contrast, the Senate may 

not place conditions on its consent to a presidential appointment. See 3 Op. Att'y Gen. 

188 (1837) (opining that Senate could not place conditions on its consent to naval commission); 

Corwin, The President, supra note 119, at 92-93 (noting that while Senate may condition 

approval of treaties, its role in relation to appointments is limited to affirming or 

rejecting nominations without condition); Henkin, supra note 9, at 122, 180-81 (same).



[Vol. 73:1791 

power in the supreme officer responsible for the faithful execution 

of the laws, to displace a subaltern officer employed in the execution 

of the laws; and a power to make treaties and to declare war, 

such as these have been found to be in their nature, their operation, 

and their consequences. 181 

Indeed, the whole notion of congressional usurpation of an executive 

function is out of place in this context. The interchangeability 

doctrine in no way restricts the President's functions; it simply gives 

him another option he can invoke when he deems it appropriate. 182 If 

there is usurpation afoot, it is not, as in the appointments context, the 

President's prerogatives that are at risk, but the Senate's. Yet, 

whether we conceive the potential encroachment as coming from the 

House or the President, the result is the same: The overwhelming dis- 

181 Helvidius No. 1, supra note 69, at 149-50. Corwin appears to have shared Madison's 

view on this point. See Corwin, President's Control, supra note 119, at 29-30 (acknowledging 

force of Madison's argument on this point). Washington also distinguished between 

the nature of the appointments and treaty powers, stating in a communication to the Senate: 

"In the appointment to offices, the agency of the Senate is purely executive, and they 

may be summoned to the President. In treaties, the agency is perhaps as much of a legislative 

nature and the business may possibly be referred to their deliberations in their legislative 

chamber." George Washington, Sentiments Expressed to the Senate Committee at a 

Second Conference on the Mode of Communication Between the President and the Senate 

on Treaties and Nominations (Aug. 10, 1789), in 30 The Writings of George Washington 

377, 378 (John C. Fitzpatrick ed., 1931). 

182 Although interchangeability expands rather than contracts executive power, Congress 

might well provoke conflict were it to go a step further and attempt to restrict the 

President's authority to enter into agreements falling within his independent constitutional 

powers. Indeed, Congress and the Executive fought a heated war of words over this issue 

in the 1970s during the heroic period of congressional efforts to rein in executive unilateralism 

in foreign affairs. Congress claimed expansive authority for restrictive framework legislation 

under its Necessary and Proper Clause power to carry into execution the powers of 

the other departments of government. Compare H.R. 4438, 94th Cong. (1975) (requiring 

President to submit executive agreements to Congress for review and possible rejection), S. 

3830, 93d Cong. (1974) (similar), and S. Rep. No. 93-1286, at 4, 7 (1974) (articulating congressional 

position), with Congressional Review of International Agreements: Hearings 

on H.R. 4438 Before the Subcomm. on International Security and Scientific Affairs of the 

House Comm. on International Relations, 94th Cong. 163-200 (1976) (testimony of 

Monroe Leigh, Legal Advisor, Dep't of State, and Antonin Scalia, Assistant Attorney 

General, Office of Legal Counsel, Dep't of Justice) (opposing congressional position). Ultimately, 

not with a bang but a whimper, Congress's resolve fizzled, and it failed to adopt 

proposed legislation which would have required congressional approval of all executive 

agreements, even those falling within the President's sole powers. The important point 

here is that the interchangeability doctrine in no way implies a resolution of this conflict in 

favor of one or the other of the contending branches. Interchangeability asserts only that 

Congress may authorize the President to enter into agreements, not that it may limit him in 

the exercise of his independent constitutional authority. On the other hand, whether Professor 

Tribe's view would undermine congressional efforts to limit presidential unilateralism 

in this way is unclear, though his emphatic insistence that Congress has no power to 

approve international agreements certainly seems to suggest a pro-Executive position.



AGAINST FREE-FORPf FORMALISM 

analogies undermine the relevance of any argument from the Appointments 

Clause. 

Consider first the intrabranch struggle between the two houses of 

Congress. The fundamental separation of powers concerns that justify 

a strict rule of construction to avoid legislative infringements on the 

executive's very capacity for independence are entirely absent in this 

context. To the extent the separation of powers principle applies at 

all, it concerns the propriety of House participation in the treaty process 

given its institutional characteristics. The Senate, when acting 

separately from the House, however, is an extraordinary body in our 

constitutional system, derogating from the carefully structured bicameral 

legislative procedure the Framers deemed necessary for the protection 

of the people. 83 We ought to assume that the Framers 

excluded the House only for compelling reasons in particular contexts 

and ought carefully to avoid extending the Senate's exclusive powers 

beyond the strict necessities of the case. If anything, then, there are 

persuasive grounds for construing the Senate's powers strictly to avoid 

any further distortion of the normal legislative process. 1 8 4 

Now, consider the interbranch struggle between the Senate as 

one component of the legislative body and the Executive-or, more 

precisely, consider the prerogative of one-third plus one of the Senate 

to veto treaties approved by the majority.1 s This conflict presents the 

converse of the Appointments Clause problem-an executive usurpation, 

aided by the House, of a legislative power, the power of a minority 

to block the approval of treaties. Here again, however, as in the 

House/Senate dispute and precisely contrary to the appointments 

power, we have compelling reasons to confine the Senate minority's 

veto as narrowly as possible. Indeed, the argument for a strict rule of 

construction has even greater force in this context because the twothirds 

supermajority requirement is a derogation from the usual 

183 See United States v. Munoz-Flores, 495 U.S. 385, 394-95 (1990) (discussing importance 

of bicameralism in protecting liberties of people from legislative overreaching and 

tyranny); I.N.S. v. Chadha, 462 U.S. 919, 948-50 (1983) (same); The Federalist No. 51, at 

322 (James Madison) (Clinton Rossiter ed., 1961) (same); The Federalist No. 62, at 378-79 

(probably James Madison) (Clinton Rossiter ed., 1961) (same); Tbe Federalist No. 63, at 

385-90 (probably James Madison) (Clinton Rossiter ed., 1961) (same). 

184 See Chadha, 462 U.S. at 955-56 (noting that exceptional cases in which one house 

may act independently of other house are "narrow, explicit, and separately justified"); 

Henkin, supra note 180, at 652-53 (arguing that no unenumerated powers should be inferred 

for Senate acting alone); cf. The Head Money Cases, 112 U.S. 5S0, 598-99 (1884) 

(suggesting that if Supremacy Clause were absent from Constitution, lack of participation 

of House in treatymaking would militate in favor of according superior status to laws). 

185 What is really at stake is not the prerogative of the Senate itself, since a majority of 

the Senate must still approve any congressional-executive agreement.


1852 


[Vol. 73:1791 

majoritarian principles of the Constitution 86 and runs the everpresent 

risk of subjecting the wishes of the majority to the will of the 

minority. 187 We ought to be particularly skeptical of any reading that 

186 Of course, bicameralism is itself a modification of a purely majoritarian system. In 

any case, there are four other instances in which the Constitution requires supermajority 

votes, all of which involve the absence of one or more of the three usual lawmaking institutions. 

See supra notes 69-73 and accompanying text. In addition, in each case there are 

specialized considerations that seem to underlie the Framers' adoption of an extraordinary 

voting requirement. 

The first is the adoption and amendment of the Constitution itself. See U.S. Const. 

art. VII (requiring ratification by nine states); id. art. V (requiring, alternatively, congressional 

amendment proposals to be adopted by two-thirds of both houses and approved by 

three-quarters of states or calling of a convention by Congress upon application of twothirds 

of state legislatures and approval of proposed amendments by three-quarters of 

states, and prohibiting any amendment depriving a state of its equal suffrage in Senate 

without its consent). Without entering into the amendments debate, it seems evident that 

the Framers held the view, widely shared even today, that there is something unique in the 

nature of a constitution such that its adoption and/or amendment should be supported by a 

strong majority of the population. 

The second case is the power to try impeachments, which, in the absence of the House 

and President, requires a two-thirds vote in the Senate for conviction. See id. art. I, § 3, cl. 

6. Since impeachment is in the nature of a criminal charge, it seems unremarkable that the 

Framers required a qualified majority of the triers of fact to be persuaded. Additionally, 

there is the need to prevent those holding a majority of seats from abusing the minority 

through political uses of the impeachment power. 

Third, each House may expel a member only upon a two-thirds vote. See id. art. 1, 

§ 5, cI. 2. Here, again, there is not only the criminal analogy but the need to prevent 

obvious abuses of majoritarian democracy. Cf. Powell v. McCormack, 395 U.S. 486,547-48 

(1969) (considering two-thirds requirement for expulsion). 

Finally, the fourth case is the veto override provision, which allows two-thirds of both 

houses of Congress to pass a law over the objections of the President. See U.S. Const. art. 

I, § 7, cls. 2, 3. In part, the President's veto power enables him to guard his own institutional 

position against an overreaching legislature and to protect the minority against 

majoritarian excesses. In this context, the President's role is comparable to that of the 

Supreme Court in exercising the power of judicial review, except that the Framers allowed 

the President to be overruled by two-thirds of both houses (whereas the Court must be 

overruled not only by two-thirds of both houses, but three-quarters of the states as well). 

In part, when the President vetoes a bill, he throws in doubt not only the constitutionality 

and/or the wisdom of the measure, but the assumption that Congress has faithfully executed 

the will of the majority. As the Court in Myers put it: 

The President is a representative of the people just as the members of the 

Senate and of the House are, and it may be, at some times, on some subjects, 

that the President elected by all the people is rather more representative of 

them all than are the members of either body of the Legislature whose constituencies 

are local and not countrywide .... 

Myers v. United States, 272 U.S. 52, 123 (1926). Hence, requiring two-thirds of both 

houses for an override helps to ensure that the law passed actually represents the majority's 

considered views. See Chadha, 462 U.S. at 948. 

187 See 2 Records of the Federal Convention of 1787, at 540 (Max Farrand ed., 1911) 

[hereinafter Records of the Convention] (remarks of James Wilson) (objecting to twothirds 

rule because it "puts it in the power of a minority to controul the will of a majority"); 

The Federalist No. 75, at 453 (Alexander Hamilton) (Clinton Rossiter ed., 1961) (noting 

that "all provisions which require more than the majority of any body to its resolutions



1853 

would expand the reach of this extraordinary requirement beyond 

what is compelled by the language of the text or is necessary to preserve 

it from being rendered nugatory. 

Regardless of how one thinks these separation of powers conflicts 

should be resolved, 18 8 what should be clear beyond dispute is that the 

Appointments Clause deals with an entirely separate problem and can 

offer no real guidance in this radically different context. The textual, 

structural, and architectural reasons for an exclusive reading of the 

Appointments Clause simply have no bearing on the proper interpretation 

of the Treaty Clause. As a consequence, we cannot sensibly 

look to this neighboring provision to help us with the difficult interpretive 

questions that the Treaty Clause presents. 18 9 

have a direct tendency to embarrass the operations of the government and an indirect one 

to subject the sense of the majority to that of the minority"). 

188 My larger point, of course, is that they cannot be resolved on the basis of text alone. 

Take the House/Senate conflict, for example. From a purely textual perspective, the question 

is whether the apparent (though necessarily speculative) institutional reasons for excluding 

the House-the large number of representatives and their short terms of office, see 

supra text accompanying notes 68-69-require their exclusion in all cases, or whether it is 

sufficient to allow the President the option of avoiding the House in those instances when 

he believes that these institutional characteristics might prejudice his foreign policy initiatives, 

say, because of the need for secrecy and dispatch. See supra notes 68-74 and accompanying 

text. The text alone does not aid in resolving the ambiguity, although, as I argue 

above, there are persuasive grounds for narrowly construing exceptions to the usual bicameral 

legislative procedure. Similarly, consider the President/Senate minority conflict. From 

a textual perspective, the question is whether the Framers empowered the Senate minority 

in order to provide a sectional veto over treaties or as an added protection in those cases 

where the President concludes that the House cannot safely participate. Again, the constitutional 

text does not resolve the issue, but, as the text above suggests, there are persuasive 

grounds here as well for strictly construing the scope of extraordinary supermajority voting 

requirements, such as the two-thirds rule. 

189 Another closely-related structural difference that undermines the effort to import 

the exclusivity of the Appointments Clause to the Treaty Clause is the contrast between 

the simple majority rule for appointments and the supermajority requirement for treaties. 

In both cases, the reasons for the choice of the Senate, at least insofar as one can infer from 

the text alone, appear similar. The Senate, among other things, would ensure a more equitable 

geographical distribution of appointments, see Myers, 272 U.S. at 119-20, and likewise 

would ensure that the interests of all states be considered equally in undertaking 

international commitments. However, the Framers' reasons for excluding the House appear 

quite different in the two contexts. In the case of appointments, at least one persuasive 

textual inference is that they wished to avoid placing any further obstacles in the 

President's path. Having already introduced the legislature into a core executive function, 

the Framers sought to avoid compounding the problem by requiring the President to run 

the gauntlet not just of the Senate but of the differently configured House as well. Ihis, in 

turn, provides a strong structural argument for reading the Appointments Clause to prohibit 

any congressional involvement beyond that expressly granted to the Senate. In contrast, 

the two-thirds supermajority requirement for treaties suggests a different basis for 

the House's exclusion. Given the extraordinary burden placed upon the President in seeking 

the Senate's consent (as well as the Framers' view that approving agreements is a legislative 

function, see infra notes 246-58 and accompanying text), there is little basis for 

inferring that they excluded the House to avoid placing an additional obstacle on the



This brings us, then, to Professor Tribe's proximity argument, his 

claim that the two clauses should be read in parallel because they appear 

in the same clause and even sentence of the Constitution. Professor 

Tribe appears to find this logic particularly compelling. 

Reinvoking his mathematical theme, he likens the argument to a virtual 

axiomatic proof. Suppose someone makes the argument that the 

age eligibility clause for Senators should be read in base eight. 190 "It 

would," he then points out, "be a fatal objection to the base eight view 

if the numeral '9' were used where the very same sentence adds that a 

Senator must have been a United States citizen for at least nine 

years."' 

91 

QED. Apparently, in his view, a similar "QED" follows 

directly from the exclusivity of the appointments power, refuting the 

interchangeability doctrine. 192 What are we to make of this argument 

for importing the physically adjacent Appointment Clause's exclusivity 

to the functionally distant Treaty Clause 

President. Limiting the burden on the President in the one case was apparently not the 

point in the second, and the reasons that underwrite a narrow, exclusive reading of the 

Appointments Clause simply do not apply to the Treaty Clause. See infra note 247 and 

accompanying text. For this reason as well, the interpretation of the Appointments Clause 

does not shed any light on the proper interpretation of the Treaty Clause. 

190 That clause requires senators to be 30 years of age. See U.S. Const. art. I, § 3, cl. 3. 

To make the hypothetical work, Professor Tribe asks us to assume that the text uses numerals 

rather than spelling out, as it does, the words "nine" and "thirty." See Tribe, Taking 

Text, supra note 3, at 1274. 

191 Id. at 1274 (citing U.S. Const. art. I, § 3, cl. 3). 

192 In a footnote, Professor Tribe concedes that the Appointments Clause argument 

does not provide "a fail-safe proof of the exclusivity of the Treaty Clause procedure." Id. 

at 1274 n.181. Perhaps, then, his hypothetical is designed just to demonstrate the importance 

for constitutional interpretation of reading whole sentences rather than just fragments. 

After all, the latter part of a sentence might reflect on the proper interpretation of 

the first part. This point, however, is so obviously uncontroversial it seems unlikely he 

would have belabored it with an elaborate hypothetical illustration. On the other hand, he 

does charge Professor Ackerman and me with deriding his suggestion that the Appointments 

Clause "should be considered in deciding the issue of the Treaty Clause's exclusivity," 

id. at 1273-74, and uses this as an occasion to provide a lecture on the importance of 

reading sentences as a whole. Needless to say, Professor Ackerman and I have not criticized 

him for "considering" the Appointments Clause. Every provision in the Constitution 

that might shed light on the interpretation of the Treaty Clause, including the Appointments 

Clause, ought obviously to be consulted, and spreading the net as widely as possible 

is laudable. The Appointments Clause in particular is a good candidate for inclusion because 

of the linguistic and formal structural similarities to the Treaty Clause. We simply 

argued that having consulted and analyzed the Appointments Clause, it has no real bearing 

on interpreting the Treaty Clause. Everything should be consulted, but arguments should 

be based only on clauses that turn out, on examination, to be relevant. Indeed, a slight 

variation on Professor Tribe's base eight story illustrates our point: Suppose that Senators 

need be citizens only seven, not nine years. Then reading the sentence as a whole, in 

Professor Tribe's sense, would be unhelpful. The latter portion would not give us any further 

evidence as to whether base eight, nine, or ten were intended, and it would not be 

worth pointing to as an argument for the base ten view. That is exactly the status of the 

Appointments Clause in our view.




Surely Professor Tribe must agree that the Constitution's grants 

of power are too nuanced, too idiosyncratic to lend themselves to simple 

mechanical formulas of this kind. Why, then, abandon his own 

past subtle efforts at constitutional elucidation in favor of arid formalism 

on stilts Consider the untenable implications of his interpretive 

theory. Recall that Article II, Section 2, Clause 2, the home of the 

Treaty and Appointments Clauses, contains only a portion of the powers 

granted to the President. Clauses 1 and 3 of that section, inter alia, 

contain additional grants. Does Professor Tribe's interpretive method 

apply to all grants contained within the same section We would then 

have to conclude that each of these powers is exclusive because the 

appointments power in Clause 2 is exclusive. 193 Or perhaps the argument 

is limited to powers that are contained in the same clause, perhaps 

only to those, like the appointments and treaty powers, that 

appear in the same sentence. The grants in Clause 1, as in most other 

clauses in the Constitution, fit this same sentence rule. They include 

the President's power as Commander-in-Chief, 194 the power to require 

the written opinions of the principal officers in each department 

"upon any Subject relating to the Duties of their respective Offices,"' 

195 and the power to grant pardons and reprieves. 196 Even giving 

it the narrowest construction, then, Professor Tribes methodology 

would require that they all be given a parallel construction-they are 

all either exclusive, nonexclusive, or mixed in the same degree, but 

not a dissimilar admixture of the three. If we take long-accepted 

views seriously, however, this cannot be right. 

Consider the Commander-in-Chief power. Only the President 

can direct troop movements, form military strategies, order a battlefield 

attack, and so on. 197 Outside this robust core of exclusivity, pres- 

193 Alternatively, returning to Professor Tribe's textual argument, are we to conclude 

that these powers are all exclusive because none of them contains any language of nonexclusivity, 

as does the proviso See supra notes 157-61 and accompanying text. 

194 See U.S. Const. art. 1, § 2, cl. 1. 

195 Id. 

196 See id. 

197 See Ex parte Miligan, 71 U.S. (4 Vall.) 2, 139 (1866) ("[The war] power necessarily 

extends to all legislation essential to the prosecution of war with vigor and success, except 

such as interferes with the command of the forces and the conduct of campaigns. That 

power and duty belong to the President as commander-in-chief."); see also Berdahl, supra 

note 118, at 116-30 (describing President's wide-ranging and exclusive power to conduct 

wartime military operations); Corwin, The President, supra note 119, at 262-63, 293-94 

(same); cL Fleming v. Page, 50 U.S. (9 How.) 603, 615-16 (1850) (holding that custom 

house in conquered territory was validly established under President's authority as 

Commander-in-Chief where duties imposed were part of military strategy to burden commerce 

with enemy); Tribe, Constitutional Law, supra note 11, § 4-4, at 220 & n.6 (acknowledging 

President's exclusive authority over military policy decisions, though locating 

source of authority in President's "inherent" power over foreign policy rather than Corn-


1856 


[Vol. 73:1791 

idents have also successfully achieved a penumbra of additional 

authority that is shared with Congress. This penumbra, in turn, has 

expanded with each of the successive great internal and external, hot 

and cold wars of our history. 198 In contrast, the pardon power is apparently 

entirely nonexclusive. 199 Although Congress may not 

abridge or limit the President in the exercise of the power, 200 over a 

century ago the Supreme Court held that despite the pardon power, 

Congress's (implied) authority extends to granting amnesties and that 

any "distinction between amnesty and pardon is of no practical importance." 

'201 More complicated is the power to require written opinions. 

From the very outset, Congress has claimed the implied power to require 

written reports from the heads of departments on any subject 

pertinent to its legislative responsibilities. 2° 2 Presidents have somemander-in-Chief 

Clause). Henkin is less certain but says "[ilt would be unthinkable for 

Congress to attempt detailed, tactical decision, or supervision, and as to these the 

President's authority is effectively supreme." Henkin, supra note 9, at 103-04. 

198 See Corwin, The President, supra note 119, at 263-93 (describing historical expansion 

of President's commander-in-chief power); Henkin, supra note 9, at 45-50, 97-112 (same). 

199 Acting under its Article I, Section 8 powers, Congress may vest the pardon power in 

executive officers other than the President, see The Laura, 114 U.S. 411, 414-16 (1885) 

(upholding power of Congress to vest in Secretary of Treasury power to remit fines, penalties, 

and forfeitures for violations of laws of United States against claim that pardon power 

is exclusively President's), or it may grant amnesty for any offense against the United 

States. "Although the Constitution vests in the President 'power to grant reprieves and 

pardons for offences against the United States, except in cases of impeachment,' this power 

has never been held to take from Congress the power to pass acts of general amnesty ...... 

Brown v. Walker, 161 U.S. 591, 601 (1896) (quoting U.S. Const. art. II, § 2, cl. 1). 

200 See United States v. Klein, 80 U.S. (13 Wall.) 128, 147-48 (1871) (striking down congressional 

attempt to limit scope of pardon power); Ex parte Garland, 71 U.S. (4 Wall.) 

333, 380 (1866) (holding that "Congress can neither limit the effect of [the President's] 

pardon, nor exclude from its exercise any class of offenders"). 

201 Brown v. Walker, 161 U.S. at 601. The difference, the Court said, was "'one rather of 

philological interest than of legal importance.' 'Amnesty' is defined by the lexicographers 

to be an act of the sovereign power granting oblivion, or a general pardon for a past offence, 

and is rarely, if ever, exercised in favor of single individuals ...." Id. at 601-02 

(quoting Knote v. United States, 95 U.S. 149, 152-53 (1877)); see also Burdick v. United 

States, 236 U.S. 79, 94-95 (1915) (describing further incidental differences between amnesties 

and pardons); Nix v. James, 7 F.2d 590, 593-94 (9th Cir. 1925) (holding that Probation 

Act does not encroach on President's pardon power). Despite the holdings of Brown v. 

Walker and The Laura, however, one might still doubt whether there really are no limits 

on Congress's powers to grant amnesty in individual cases or to vest a pardon power concurrently 

in other executive officers. 

M2 See Act of Sept. 2, 1789, ch. 12, 1 Stat. 65, 66 (establishing Treasury Department and 

requiring Secretary of Treasury "to make report, and give information to either branch of 

the legislature, in person or in writing (as he may be required), respecting all matters referred 

to him by the Senate or House of Representatives, or which shall appertain to his 

office"). For a very pro-Executive recounting of the early history, see Subcommittee on 

Constitutional Rights of the Senate Comm. on the Judiciary, 85th Cong., The Power of the 

President to Withhold Information from the Congress-Memorandums of the Attorney 

General pt. 1, at 2-42, 74-82 (Comm. Print 1958 & 1959) [hereinafter Attorney General's




times denied that power and, in any case, have on occasion asserted 

executive privilege to prevent the revelation of information they 

deemed confidential. 203 As a practical matter, though, presidents 

Memorandums]. But cf. H.R. Rep. No. 86-2207 (1960) (describing historical conflicts over 

executive privilege). 

Although Congress's investigatory power, like its power to approve international 

agreements, is not mentioned in the text, it has long been recognized as an implied power 

incident to the power to legislate and includes the power to compel testimony through 

subpoena. See, e.g., McGrain v. Daugherty, 273 U.S. 135,160-75 (1927) (upholding powers 

of House and Senate to compel testimony of witnesses when needed to enable them to 

perform their legislative functions); see also James M. Landis, Constitutional Limitations 

on the Congressional Power of Investigation, 40 Harv. L Rev. 153 (1926) (recounting early 

history of congressional exercises of its investigatory powers). The issues take on some 

added separation of powers complications when Congress wishes to compel the testimony 

of heads of the executive departments or to require them to make reports. See Tribe, 

Constitutional Law, supra note 11, § 4-16 (describing interplay between congressional investigatory 

power and executive privilege to withhold information). The Supreme Court 

has suggested that the power to require reports is incident to the legislative function. See 

Morrison v. Olsen, 487 U.S. 654, 694 (1988) (indicating that receiving reports from executive 

officers is "incidental to the legislative function"). The real issue is the existence and 

scope of executive privilege and its impact on Congress's implied investigatory power. 

Thus far, the Supreme Court has managed largely to stay out of the fray, although in 

United States v. Nixon, 418 U.S. 683 (1974), it did recognize the constitutional status of 

executive privilege but ruled that the courts could overrule the President's judgment at 

least in the context of a criminal prosecution. The Court left open whether executive privilege 

could be invoked in a legislative setting. See id. at 712 n.9. In any case, Congress has 

always required principal officers of the executive departments to provide written reports 

on subjects of congressional interest. Although there was some early confusion, see 

Thomas Jefferson, Diary Entries (Mar. 12 & Apr. 2, 1792), in 4 Memoir, Correspondence, 

and Miscellanies, from the Papers of Thomas Jefferson 463-65 (Thomas Jefferson 

Randolph ed., Charlottesville, F. Carr & Co. 1829) (revealing early misgivings of first cabinet); 

William Rawle, A View of the Constitution of the United States 171-72 (Philadelphia, 

Philip H. Nicklin 2d ed. 1829) (describing practice); 3 Joseph Story, Commentaries on the 

Constitution of the United States 371 n.1 (Boston, Hilard, Gray & Co. 1833) (describing 

early confusion but noting that practice is accepted), the practice is settled and has become 

even more widespread today than in the past. Indeed, it is impossible to glance through a 

volume of the United States Code without noticing some of the scores, even hundreds of 

statutes ordering executive branch officials to submit regular reports to Congress on every 

conceivable subject. For an example close to home, consider the elaborate report Congress 

requires the United States Trade Representative to submit annually on the subject of the 

World Trade Organization. See 19 U.S.C. §§ 3534, 3535 (1994). Perhaps the most widely 

known instance is the State Department's annual Country Reports on Human Rights Practices. 

See, e.g., U.S. Dep't of State, Country Reports on Human Rights Practices for 1995 

(Comm. Print 1996). The State Department submits these elaborate reports in accordance 

with a number of different statutory mandates. See, e.g., 22 U.S.C. § 2304(b) (1994 & 

Supp. 1996) (prescribing content of reports); see also U.S. Dep't of State, supra, at ix (noting 

statutory compliance). There are countless other examples. By statute, Congress has 

charged the standing committees of both houses with evaluating the execution of the laws 

by the executive branch and authorized them to "require a Government agency to [carry 

out the necessary analysis, appraisal, and evaluation themselves] and furnish a report 

thereon to the Congress." 2 U.S.C. § 190d(a) (1994). 

2M See the comprehensive listing of precedents in Attorney General's Memorandums, 

supra note 202, pt. 1, at 30-32; id. pt. 2, at 120-21; see also Corwin, The President, supra



have generally allowed department heads to provide demanded reports, 

and controversy has swirled only over the relatively few instances 

when presidents have refused to cooperate. 2 4 What is most 

striking for our purposes is that presidents, attorneys general, and 

their academic boosters have not even thought the President's power 

to require opinions worth mentioning in arguing for the Executive's 

right to withhold information from Congress, resting instead on general 

principles. 205 

Even were it the case, however, that these three powers had been 

given a roughly parallel construction, it would be nothing more than 

sheer coincidence. Given their widely different purposes and functions, 

there is no persuasive reason why they ought to be construed 

together. 20 6 It makes no difference at all that the Framers, for 

whatever reasons, grouped these three powers in the same article, section, 

clause, and sentence of the Constitution. For these reasons, it is 

not surprising, though it is quite revealing, that courts and commentators 

have not thought to consider either of the two other powers when 

construing the third. 207 Indeed, Professor Tribe himself is no exception.20 

The same principle applies to the treaty and appointments 

note 119, at 125-33, 212-13, 473-76 (discussing precedents); Henkin, supra note 9, at 115-16 

(same). 

24 In fact, most of the controversy has been over presidential refusals to produce 

records, rather than refusals to make reports. It has not been unusual for one or the other 

house of Congress to decide that to assess the accuracy of executive branch reporting, it is 

necessary to have access to the underlying records and not just to formal reports and opinions. 

See Attorney General's Memorandums, supra note 202, pt. 1, at 4-30 (citing instances); 

id. pt. 2, at 91-121 (same); Corwin, The President, supra note 119, at 126-31 

(same). 

205 At least so far as my admittedly less than complete historical research reveals. See, 

e.g., Attorney General's Memorandums, supra note 202, pt. 1, at 1-4, 69-72 (resting on 

separation of powers grounds rather than President's power to require opinions); 40 Op. 

Att'y Gen. 45 (1941) (resting on separation of powers and protection of public interest); 

Roman L. Hruska, Executive Records in Congressional Investigations-Duty to Disclose-Duty 

to Withhold, 35 Neb. L. Rev. 310 (1955) (resting on executive prerogative and 

need to protect public interest). Professor Tribe never mentions this power in his discussion 

of executive privilege. See Tribe, Constitutional Law, supra note 11, §§ 4-15 to 4-16. 

206 It is hardly plausible, for example, to suggest that we can substantially advance our 

thinking about conflicts between the political branches over the war powers by examining 

the pardon power. Nor does the status of the power to require written opinions seem to 

have any bearing on whether Congress may limit the President from the use in war, say, of 

certain kinds of weapons, or on whether a congressional amnesty for a named individual 

ought to be upheld. 

207 Compare Ex parte MiUlligan, 71 U.S. (4 Wall.) 2 (1866) (construing commander-inchief 

power), Corwin, The President, supra note 119, at 262-63,293-94 (same), and Henkin, 

supra note 9, at 45-50, 97-112 (same), with Brown v. Walker, 161 U.S. 591, 601-02 (1896) 

(construing pardon power). 

208 See Tribe, Constitutional Law, supra note 11, § 4-7 (construing Commander-in-Chief 

Clause); id. § 4-11 (construing pardon power); id. §§ 4-15 to 4-16 (construing scope of executive 

privilege).



1859 

powers-even though they inhabit the same constitutional sentence. 

Professor Tribe's same sentence rule would certainly make constitutional 

law easier, but constitutional interpretation, thankfully, is more 

subtle than that! 20 9 

2 Nor would matters improve were we to apply Professor Tribe's methodology outside 

Article II, say, to Article L The latter introduces a new worry-we have to determine not 

only whether the powers granted are exclusive of the executive but also whether they are 

exclusive of the legislative powers of the states. See supra note 115 (discussing exclusivity 

vis-a-vis states); infra notes 278-79 and accompanying text (same). To compound matters, 

all of the grants contained in Section 8 are contained within a single sentence! In the 

Constitution a la Tribe, this would appear to mean that all should be given the same construction 

in these two respects. Thus, for example, as to exclusivity vis-a-vis the states, 

compare Congress's power to borrow money on the credit of the United States, see U.S. 

Const. art. I, § 8, cl. 2, to establish post offices and post roads, see id. art. I, § 8, cl. 7, and to 

constitute tribunals inferior to the Supreme Court, see id. art. I, § 8, c. 9, with its power to 

lay and collect taxes, duties, imposts, and excises, see id. art. I, § 8, cl. 1, to regulate commerce 

with foreign nations and among the several states, see id. art. I, § 8, cl. 3, and to 

promote the progress of science and the useful arts by securing for limited times to authors 

and inventors the exclusive rights to their respective writings and discoveries, see id. art. 1, 

§ 8, cL. 8. While the former seem clearly exclusive, the latter, however different in detail, 

are nonexclusive to a significant extent (although congressional supremacy, in the event of 

conflict, is clear at least as to the latter two). See Goldstein v. California, 412 U.S. 546,552- 

60 (1973) (holding that copyright power is concurrent but that federal legislation controls 

in event of conflict); Tribe, Constitutional Law, supra note 11, §§ 6-2 to 6-5 (describing 

Court's negative Commerce Clause jurisprudence); see also The Federalist No. 32, supra 

note 94, at 199 (arguing for nonexclusivity of taxation power as well as its nonhierarchical 

character). Similarly, the grants in Article I cannot be read as uniformly exclusive of executive 

power. In this context, compare Congress's power to lay and collect taxes, see US. 

Const. art. I, § 8, cl. 1, and to borrow money on the credit of the United States, see id. art. 

I, § 8, cl. 2, and to provide for the punishment of counterfeiting the securities and current 

coin of the United States, see id. art. I, § 8, cl. 6, with its power to make rules for the 

government and regulation of the land and naval forces, see id. arL I, § 8, cl. 14, and to 

provide for organizing, arming, and disciplining the militia, see id. art. I, § 8, cL 16, and to 

regulate commerce with foreign nations, see id. art. I, § 8, cl. 3. Again, the former are 

clearly exclusive, while the latter present more complexities. As to presidential intrusions 

on Congress's power to make rules for the regulation of the land and naval forces, see 

supra note 120. As to invasions of the power to regulate foreign commerce, see supra note 

120. As to presidential overlaps with a variety of congressional powers through the conclusion 

of treaties and unilateral executive agreements, see supra note 120 and accompanying 

text. 

If we were more generous to Professor Tribe, we might assume that be would limit his 

presumption only to powers listed in a single clause-the same sentence rule being a necessary, 

but not a sufficient condition, to the operation of his presumption. But this too is of 

little help, as careful consideration of the relevant provisions reveals. The Framers were 

far more careful in Article I, Section 8 than in Article 11 to group together into single 

clauses powers that are similar in their nature and purposes. They were also more reluctant 

to group powers at alL Nevertheless, the single clause rule quickly runs into difficulties 

in this Article as well. Take the power to establish a uniform rule of naturalization and 

uniform laws on the subject of bankruptcies. See U.S. Const. art. I, § 8, cL 4. The requirement 

of national uniformity in both cases might well have led the Court to conclude that 

both powers are exclusive of the states. In fact, however, the Court early held that the 

bankruptcy power was concurrent and that state laws on the subject would not ba preempted 

unless Congress occupied the field (which, in fact, it did not for many decades).



[Vol. 73:1791 

3. Having More Fun with Word Games: Congress's Less than 

Usual Legislative Powers and the Compact Clause 

Formalism is rarely far beneath the surface in Professor Tribe's 

interpretive methodology, and it resurfaces once again in his second 

crack at the argument from expressio unius. He points out that the 

Constitution grants Congress a number of powers in provisions not 

found in Article I, Section 8. Some of these, moreover, are not "normal 

lawmaking" powers. 210 "A serious textual analysis cannot simply 

See Stellwagen v. Clum, 245 U.S. 605, 615 (1918); Sturges v. Crowninshield, 17 U.S. (4 

Wheat.) 122, 196 (1819). The naturalization power proved more difficult to interpret. An 

early case seemed to suggest that the naturalization power was also nonexclusive, leaving 

the states free in the absence of contrary federal legislation to naturalize citizens. See 

Collet v. Collet, 6 F. Cas. 105, 106-07 (C.C.D. Pa. 1792) (per curiam). But later cases, 

understandably, criticized this view. See, e.g., The Passenger Cases, 48 U.S. (7 How.) 283, 

556 (1849) (Woodbury, J., dissenting) (citing cases); The License Cases, 46 U.S. (5 How.) 

504, 585 (1847) (opinion of Taney, C.J.), overruled on other grounds by Leisy v. Hardin, 

135 U.S. 100 (1890); Chirac v. Chirac, 15 U.S. (2 Wheat.) 259, 269 (1817) (Marshall, C.J.). 

Even justices who argued for a strong presumption in favor of concurrent state jurisdiction 

believed the naturalization power exclusive. Chief Justice Taney thought it "hardly consistent" 

with the constitutional structure "to allow any one State, after the adoption of the 

constitution, to exercise a power, which, if it operated at all, must operate beyond the 

territory of the State, and compel other States to acknowledge as citizens those whom it 

might not be willing to receive." The License Cases, 46 U.S. (5 How.) at 585. Likewise, 

Chief Justice Marshall concluded: "That the power of naturalization is exclusively in congress 

does not seem to be, and certainly ought not to be, controverted ... ." Chirac, 15 

U.S. (2 Wheat.) at 269. 

Consider as well the power to declare war, grant letters of marque and reprisal, and 

make rules concerning captures on land and water. See U.S. Const. art. I, § 8, cl. 11. Narrowly 

construed, the power to declare war is exclusively vested in Congress. No President 

has ever claimed the right to issue a formal declaration of war. See Henkin, supra note 9, 

at 97. In a more extended sense, however, the war powers of the two branches are substantially 

overlapping. As Henkin puts it, "[n]o one ... can disentangle the war powers of the 

two branches." Id. at 94; see also id. at 96-112 (elaborating on overlap). In contrast, the 

now obsolete power to grant letters of marque and reprisal might well be limited to a 

formal power to issue documents having a special significance in international law circa 

1787 (that is, to save privateers from punishment as pirates under the law of nations). But 

see Jules Lobel, "Little Wars" and the Constitution, 50 U. Miami L. Rev. 61, 66-70 (1995) 

(arguing that marque and reprisal power implicitly granted Congress exclusive authority 

over limited wars and hostilities short of formal war). If the more limited view is right (and 

perhaps even if the broader view is correct), the power is almost surely exclusively lodged 

in Congress. Finally, contrast the power to make rules concerning captures. In an early 

case, the Court, over Justice Story's strong dissent, held that the President had no power to 

order the confiscation of alien enemy property during war. That power was lodged in 

Congress. See Brown v. United States, 12 U.S. (8 Cranch) 110, 128-29 (1814). During the 

Civil War, the Court seemed to abandon this view. See The Prize Cases, 67 U.S. (2 Black) 

635, 670-71 (1862) (holding that under law of war President may, without congressional 

authorization, order blockade and capture of vessels that violate blockade). And commentators 

have strongly doubted whether the Court would today interfere with presidential 

confiscations of enemy property during war, even in the absence of congressional legislation. 

See, e.g., Henkin, supra note 9, at 104. Perhaps, then, the power is concurrent. 

210 Tribe, Taking Text, supra note 3, at 1269.




slide past the brute fact that international agreement-making on behalf 

of the United States is not among these enumerated powers." 211 

To be sure, Article I, Section 8 is not the sole repository of Congress's 

powers, and some of the powers found outside that section arguably 

fall outside some narrow conception of "normal" 

lawmaking. 212 But this is equally true of the powers enumerated in 

Section 8,213 and it is difficult to see how pointing to these additional 

powers adds anything but emphasis to the argument. Why would the 

fact that some of those powers appear outside of Article I, Section 8 in 

any way strengthen the expressio unius inference 214 

Furthermore, when we examine the particular powers to which 

Professor Tribe refers, his argument appears even less justified. Some 

are simply "normal" lawmaking powers or scarcely distinguishable 

from them. 215 The bulk of the others are found in Article I, Section 

10, Clauses 2 and 3. These set forth a list of powers the states are 

forbidden from exercising except with the consent of Congress. 21 6 

Professor Tribe seems to think it particularly significant that in these 

cases Congress is given an express power of "consent." In his view, 

this indicates that the Framers would have been explicit had they 

211 Id. at 1270. In Professor Tribe's view, of course, approving agreements is not a "normal 

lawmaking" function. For an extended reply to this assumption, implicit in the cited 

passage, but made explicit elsewhere, see infra notes 24462 and accompanying text. 

212 Professor Tribe cites Article I, Section 9, Clause 8 (prohibiting any person holding 

any office under United States from accepting any present, emolument, office, or title from 

foreign state without consent of Congress); Article I, Section 10, Clause 2 (prohibiting 

states from taxing imports or exports without consent of Congress); Article 1, Section 10, 

Clause 3 (prohibiting states, without consent of Congress, from laying any duty of tonnage, 

keeping troops or ships of war in time of peace, entering into any agreement or compact 

with another state or with foreign power, or engaging in war unless actually invaded or in 

imminent danger); Article IV, Section 1 (granting Congress power to regulate public 

records); Article IV, Section 3, Clause 1 (granting power to admit new states or consent to 

formation of new states from other states or parts thereof); Article IV, Section 3, Clause 2 

(granting power to make all needful rules and regulations respecting territory or property 

of United States); and Article V (granting Congress power to propose constitutional 

amendments by two-thirds vote in each house). See Tribe, Taking Text, supra note 3, at 

1270 & n.170. 

213 Consider, for example, Article I, Section 8, Clause 5 (granting Congress power to 

coin money), Clause 11 (granting power to declare war and to grant letters of marque and 

reprisal), and Clause 12 (granting power to raise and support armies). 

214 The stronger expressio unius argument rests on the grant of the treaty power to the 

President and the Senate, not on the nature of the expressly enumerated powers granted to 

Congress. McCulloch's expansive reading of the Necessary and Proper Clause undermines 

the force of the latter. See supra notes 49-56 and accompanying text; infra notes 228-31, 

235-38, and accompanying text. Of course, I consider the former weak as well. See supra 

notes 93-100, 116-23, and accompanying text. 

215 See, e.g., U.S. Const. art. IV, § 1 (granting power to regulate public records); id. art. 

lV, § 3, cl. 2 (granting power to make all needful rules and regulations for territory and 

property of United States). 

216 See supra note 212.



[Vol. 73:1791 

wished Congress to have the power to "consent" to federal agreements. 

217 It should be clear, of course, that this is just another highly 

formulaic version of the expressio unius argument-because the 

Framers granted Congress the power of "consent" in some cases, they 

must have meant to exclude the power of "consent" in all other cases. 

Not only is this approach overly mechanical, it becomes virtually indefensible 

once we view the relevant provisions in context. 

The Framers expressly referred to congressional consent only because 

they wished to deny certain powers to the states but nevertheless 

to permit them to act under leave of Congress. It was thus 

necessary to dispel any inference arising from the denial of these powers 

that the bar was absolute, beyond even Congress's power to override. 

This need was particularly pressing, moreover, because of the 

structure of Section 10. The first clause contains a list of powers flatly 

denied the states, while the second and third, on which Professor Tribe 

myopically focuses, forbid the states from exercising a further list of 

powers except with congressional consent. 218 Had the Framers deleted 

the qualifying language, two possible inferences would have 

been compelling: Either all of the powers listed were absolutely denied 

to the states, or all could be exercised but only if Congress, acting 

under the Necessary and Proper Clause, were to give its consent. It is 

manifest, however, that neither of these inferences is consistent with 

what the Framers sought to achieve-an absolute prohibition on exercising 

the powers in the first clause and a qualified prohibition on 

those in the second and third. 219 This explains, then, why they included 

the language on which Professor Tribe relies and why its inclusion 

has no bearing on whether Congress has an implied power to 

approve international agreements. The text's reference to congressional 

consent in this special context has no tendency to suggest that 

217 See Tribe, Taking Text, supra note 3, at 1270. 

218 Article I, Section 10, Clause 1 prohibits the states from entering into any treaty, 

alliance, or confederation; granting letters of marque and reprisal; coining money; emitting 

bills of credit; making any thing but gold and silver coin a legal tender; passing any bill of 

attainder, ex post facto law, or law impairing the obligation of contracts; or granting any 

title of nobility. Professor Tribe has recognized these as absolute prohibitions. See Tribe, 

Constitutional Law, supra note 11, § 6-33, at 521, 527 n.42; see also infra note 362 (discussing 

implications and rationale for structure of Section 10). For the provisions of Clauses 2 

and 3, see supra note 212. 

219 See supra note 218; infra note 362. Perhaps, the Court would have determined on a 

case-by-case basis whether a particular power included in Article I, Section 10 required 

absolute prohibition in light of its peculiar nature or whether the prohibition could be 

overcome by congressional consent. This, too, however, would not necessarily have 

achieved what the Framers had in mind-the particular division they made among the 

powers included in Clauses 1, 2, and 3.



AGAINST FREE-FORM FORALISM 

the Framers would have been express on all other occasions when 

they wished Congress to have a power of "consent."2 ' 0 

It is not surprising that Professor Tribe shifts quickly from the 

general to the specific, moving from Article I, Section 10 as a whole to 

one provision in particular, the Compact Clause. As we have seen, 

that clause empowers Congress to approve (give consent to) agreements 

negotiated by the states. This demonstrates, he thinks, that the 

Framers were explicit whenever they contemplated "congressional supervision 

of any category of agreements with foreign nations." ' 2 1 This 

"quite badly damage[s]" the case for the congressional-executive 

agreement, since, of course, the Constitution is silent on whether Congress 

has the power to approve federal agreements.m 

220 Indeed, as Professor Tribe has previously recognized, the Framers were not in fact 

express on other occasions when they did wish Congress to have a power of consent. Consider, 

for example, Congress's power to "consent" to state regulations of interstate commerce 

that would otherwise be beyond state authority because they unduly discriminate 

against out-of-state commerce. See Tribe, Constitutional Law, supra note 11, § 6-33, at 

524-27 (discussing source and scope of power). More generally, although he has recognized 

that Congress may not override the absolute prohibitions on state regulation contained 

in the first clause of Article I, Section 10, Professor Tribe has affirmed that there are 

implied congressional consent powers over state action in two broad areas which go beyond 

the express consent powers specified in the second and third clauses of Section 10. 

[C]ongressional consent or ratification may suffice to validate otherwise unconstitutional 

state action in three different settings: first, where the Constitution 

expressly makes congressional consent a prerequisite of state action, as in 

the provisions of article I, § 10, with respect to import and export duties, interstate 

compacts, and certain other topics; second, where the existence of a constitutional 

ban on state action is inferred entirely from a grant of legislative 

power to Congress, as in the case of the commerce clause; and third, where the 

constitutional prohibition against state intrusion is thought to follow from concerns 

of federalism that may properly be entrusted to Congress, as in the case 

of federal immunity from state taxation. 

Id. § 6-33, at 521. His current argument, then, appears to be wholly inconsistent with his 

previous views. 

Furthermore, the analysis in the text above applies as well to the provision prohibiting 

any person holding office under the United States from accepting any present, office, or 

title from a foreign state without the consent of Congress. See U.S. Const. art. I, § 9, c. 8. 

That provision likewise appears among a list of powers, in this case denied to the federal 

government, including those holding office under it. It would have been impossible to 

determine which activities were absolutely prohibited and which prohibited except with 

congressional consent had the Framers failed to include explicit language in those instances 

where they intended the latter. 

Professor Tribe's reliance on Article V is likewise misplaced. The need to make Congress's 

power to propose amendments explicit arose from at least two considerations. It 

was at least doubtful whether a power to amend the constituent text could be inferred from 

Congress's implied legislative powers. The Framers, moreover, vished to limit Congress's 

role to proposing amendments rather than adopting them and to require a two-thirds vote 

in both houses rather than the usual simple majority with the concurrence of the President. 


222 Id.



[Vol. 73:1791 

As with many of his other arguments, there is an embarrassing 

array of reasons why this argument fails to persuade. Begin with the 

point we just noticed. The Framers had very specific reasons for including 

the reference to congressional consent in the Compact Clause. 

In the first clause of Section 10 they had prohibited states from entering 

into treaties, alliances, or confederations, and in the third clause 

they wished to prohibit them from entering into agreements or compacts. 

Only in the latter case, as opposed to the former, they wished 

to qualify the prohibition by permitting the states to make agreements 

or compacts with congressional consent. To make this distinction they 

had to include the language on which Professor Tribe relies. Otherwise, 

the states would most likely have been barred altogether from 

making international agreements of any kind or, less likely, have been 

permitted to do so across the board but only with the permission of 

Congress. Thus, unless we think the Framers were engaged in an elaborate 

exercise in linguistic forms, we have no reason to believe that 

they included the qualifying language because of a commitment to being 

explicit whenever they contemplated "congressional supervision of 

any category of agreements." Professor Tribe's inference is extremely 

frail at best. 

Second, Professor Tribe himself seems to recognize its weakness. 

"Perhaps," he suggests, the Framers failed to grant Congress an explicit 

power to approve agreements "simply because the giving of such 

consent was understood, or tacitly assumed, to be included in Congress's 

section 8 powers. '' 2 2 This argument can be rejected, however, 

because "such consent was not thought to be so included for the first 

century-and-a-half of constitutional practice. '224 Quite so! That is 

precisely what Professor Ackerman and I have been at pains to argue: 

The text is indeterminate, but reference to practice during the first 

hundred and fifty years resolves the interpretive doubts. But if Professor 

Tribe is willing to resort to history to supplement his textual 

arguments, what exactly are we arguing about And if resort to history 

is necessary to support his textual claims, then we seem to agree 

that the text itself is indeterminate. In other words, this concession 

seems to undermine his entire argument. 

Third, Professor Tribe's peripheral vision again seems impaired, 

for in word games, turnabout is fair play. Thus, if the Compact Clause 

suggests that the Framers were explicit when they contemplated that 

Congress could approve agreements, then Article I, Section 10's explicit 

prohibition on state treaties equally suggests that they were ex- 

223 Id. at 1270. 

224 Id.



AGAINST FREE-FORM-! FORMALISM 

plicit whenever they wished to deny the power to make treaties. But 

there is no provision in the text expressly denying Congress the power 

to approve treaties. How are we to know which of these is the more 

compelling inference I would suggest that the better approach is to 

foreswear both as unworthy of a serious effort at constitutional 

exegesis. 

Finally, Professor Tribe's argument runs into further trouble 

when confronted with the unilateral executive agreement, which, as 

we will see, he is at pains to defend.72 s Even if we accept his reading 

of the Compact Clause, the inference he chooses to draw is oddly narrow. 

The lesson, it would seem, is that the Framers were explicit when 

they wished to provide for the making of agreements, not just when 

they wished to provide for congressional supervision of agreementmaking. 

However, the only provision permitting the President to 

make agreements is the Treaty Clause; hence, the unilateral agreement 

should be ruled out. Indeed, even if we accept Professor Tribe's 

crimped version of the lesson of the Compact Clause, the same problem 

arises. The Compact Clause, after all, is not the only provision for 

the making of agreements. The Treaty Clause, it will be recalled, 

grants that power to the President with the advice and consent of the 

Senate. But if Congress's power to approve state agreements suggests 

that the Framers were explicit about its agreement-making powers, 

then the Treaty Clause, by a parity of reasoning, suggests that they 

were explicit about the President's authority to make agreements. 

Yet, search as we may, the text yields only silence on the unilateral 

executive agreement.22 s 

There is considerable irony here. From the express and affirmative 

grant to Congress of the power to approve state agreements, Professor 

Tribe wishes to draw the implication that Congress is excluded 

from any further role in approving agreements. But from the absence 

of any reference to a presidential power to make agreements (outside 

the Treaty Clause), he finds an important unenumerated executive 

agreement-making power.2 7 As a thought experiment, imagine that 

the Compact Clause had granted the approval power to the President 

rather than to the Congress. Would Professor Tribe have then said 

that the President's agreement-making powers (outside the Treaty 

775 For further discussion, see infra Part III.D.2. 

226 As soon as we begin finding explanations for their silence in the one case, moreover, 

we then have to consider what reasons they may have had for silence in the other. In other 

words, once we postulate reasons for not drawing the erpressio unius inference as to the 

President's unilateral agreement-making powers, we can at least as persuasively posit reasons 

for not drawing the inference in Congress's case as well. 

227 See infra Part III.D.2.


Clause) are limited to state agreements Would he then have affirmed 

a congressional power to approve federal agreements On the 

contrary, it seems evident that exhibit number one in his brief against 

congressional power would be the grant to the President rather than 

Congress of the power to approve state agreements. This would be 

cited as evidence that the Framers wished to exclude Congress altogether 

from the agreement-making process, and any negative implication 

regarding the President's unenumerated powers would have been 

waived away on grounds like those I have already articulated. 

B. Negative Textual Arguments 

Professor Tribe makes one principal textual argument against the 

case for the congressional-executive agreement: He seeks to undermine 

the availability of the Necessary and Proper Clause as a basis for 

Congress's power to approve international agreements. I address this 

argument below, as well as two brief points he adds as grace notes. 

1. Approving Agreements as Necessary but Improper. Isn't There 

Some Way Around the Necessary and Proper Clause 

The affirmative textual case for the congressional-executive 

agreement rests fundamentally on Chief Justice Marshall's ringing affirmation 

in McCulloch v. Maryland 228 of the scope of Congress's implied 

powers under the Necessary and Proper Clause: "Let the end be 

legitimate, let it be within the scope of the constitution, and all means 

which are appropriate, which are plainly adapted to that end, which 

are not prohibited, but consist with the letter and spirit of the constitution, 

are constitutional." 229 In one fell swoop, Marshall thus swept 

away the fierce objections of his Republican adversaries Jefferson and 

Madison 230 and established a foundational construction of the fundamental 

text ever since accepted by the courts and the political 

branches. 231 It is beyond dispute that when Congress authorizes the 

228 17 U.S. (4 Wheat.) 316 (1819) (Marshall, C.J.). 

229 Id. at 421. 

230 Jefferson had argued that Congress's implied powers under the Necessary and 

Proper Clause are limited to those "without which the [express] grant[s] of power would be 

nugatory," Thomas Jefferson, Opinion on the Constitutionality of the Bill for Establishing 

a National Bank (Feb. 15, 1791), in 19 The Papers of Thomas Jefferson 275, 278 (Julian P. 

Boyd ed., 1974), and Madison had taken the view that they are limited to those "evidently 

and necessarily involved in an express power" and which are not inherently important 

powers in themselves, 2 Annals of Cong. 1899-1901 (1791) [hereinafter 2 Annals] (remarks 

of Rep. Madison). For further discussion of Madison's view, see infra note 268. For further 

commentary on the epochal dispute over the first Bank of the United States, see Paul 

Brest & Sanford Levinson, Processes of Constitutional Decisionmaking 9-17 (3d ed. 1992). 

231 Professor Tribe does not purport to revisit McCulloch. Perhaps as a strict textualist 

he should. Some have argued, not unpersuasively, that Marshall was de facto amending 

Imaged with the Permission of N.Y.U. Law Review




President to conclude an international agreement in order to implement 

its regulatory responsibilities under one of its broad foreign affairs 

powers, it has utilized a "means which [is] appropriate, which [is] 

plainly adapted" to a legitimate end "within the scope of the constitution." 

The only question, then, is whether this means is "prohibited"-whether, 

in other words, we ought to read the Treaty Clause as 

commanding that the senatorial advice and consent procedure is the 

exclusive method by which international agreements may be approved. 

Which brings us back to our initial question: Is the Treaty 

Clause exclusive If not, then the Necessary and Proper Clause supplies 

clear textual grounds for the congressional-executive agreement. 

Professor Tribe has two responses to the Necessary and Proper 

Clause. First, he points out that the mere fact that Congress adopts a 

law as a means to implement one of its enumerated powers is not 

sufficient to establish its constitutionality. Even though the law is 

"necessary and proper" to implementing Congress's express powers, it 

will still be unconstitutional if it contravenes other provisions of the 

Constitution, such as the Bill of Rights or the provisions creating 

structural constraints on Congress3P 2 Professor Tribe is, of course, 

correct so far as it goes, but his argument begs the fundamental question. 

McCulloch recognizes that the Necessary and Proper Clause 

does not authorize Congress to adopt laws that are inconsistent with 

"the letter and spirit of the constitution." Moreover, as Professor 

Tribe points out, the more recent decisions of the Court in LN.S. v. 

Chadha 33 and New York v. United States2 34 reaffirm this constitutional 

commonplace. But the mere fact that the Necessary and Proper 

Clause does not automatically insulate a law from constitutional challenge 

does not establish, or even argue in favor of, the exclusivity of 

the Treaty Clause, nor does it undermine the Necessary and Proper 

Clause as a foundation for the congressional-executive agreement. 

Whether that clause will support the congressional-executive agreement 

will depend upon how the other arguments in favor of and 

the Constitution of 1787. See Sanford Levinson, How Many imes Has the United States 

Constitution Been Amended (A) < 26; (B) 26; (C) 27; (D) > 27: Accounting for Constitutional 

Change (citing James Boyd White, When Words Lose Their Meaning 263 

(1984)), in Responding to Imperfection, supra note 17, at 13,22-23. 


233 462 U.S. 919 (1983) (striking down legislative veto on structural grounds even though 

a useful device for implementing Congress's legislative powers). 

234 504 U.S. 144 (1992) (striking down on federalism grounds statute directing states to 

regulate interstate commerce in radioactive waste in accordance with congressional instructions). 

The even more recent decision in Printz v. United States, 117 S. Ct. 2365 

(1997) (striking down on federalism grounds provisions of Brady Handgun Violence Prevention 

Act that required state police to conduct background checks of handgun purchasers), 

is to the same effect.


1868 NEW YORK UNIVERSITY LAW REVIEW [V/ol. 73:1791 

against the exclusivity of the Treaty Clause are resolved. If the nonexclusive 

view prevails, then the Necessary and Proper Clause will supply 

the essential constitutional predicate for the modem practice. 

Professor Tribe's first argument is thus beside the point; his second 

is entirely unconvincing. He points out that the Necessary and 

Proper Clause only permits Congress to "make all Laws" that are necessary 

to executing its enumerated powers23 

3 5 But, he argues, approving 

an international agreement is not making a law; it is an unspecified 

something else.2 6 It should be immediately clear that Professor Tribe 

is really urging that we revisit McCulloch. For Marshall was abundantly 

clear that the implied powers doctrine, incorporated into but 

existing independently of the Necessary and Proper Clause, authorizes 

Congress to adopt all means that are adapted to executing its express 

powers.Z 37 Approving an agreement, like chartering a bank, is surely 

a useful "means" of implementing Congress's enumerated foreign affairs 

powers and is thus well within the McCulloch framework. So too 

are making paper money a legal tender, compelling testimony, punishing 

contempts, and chartering corporations.23 None involve the passage 

of laws setting forth rules of general application, but all are surely 

within the scope of Congress's implied powers. 

In any case, it is extremely difficult to see why Professor Tribe 

doubts that approving an agreement falls under the language of the 

Necessary and Proper Clause. Perhaps it is simply the result of the 

way he frames the problem. When Congress "approves" an agreement, 

it does not engage in some international ritual unconnected to 

the legislative process, any more than when Congress created the 

Bank of the United States, it laid the bricks and mortar for its building 

or printed the corporate seal. As with any other law, Congress approves 

an international agreement through the passage of a bill or 

235 U.S. Const. art. I, § 8, cl. 18. 


237 Chief Justice Marshall uses the term "means" throughout the opinion. See 

McCulloch v. Maryland, 17 U.S. (4 Wheat.) 316 passim (1819) (Marshall, C.J.). He also 

treats the implied powers doctrine as independent of the Necessary and Proper Clause. 

See id. at 406-11 (finding implied congressional power to create corporations, and only 

subsequently discussing Necessary and Proper Clause in response to argument by counsel); 

Tribe, Constitutional Law, supra note 11, § 5-3, at 301 n.5 (stating that "Marshall, like 

Madison, did not regard the necessary and proper clause as the source of the implied congressional 

power recognized in McCulloch"). For Madison's concurring view, see The Federalist 

No. 44, at 285 (James Madison) (Clinton Rossiter ed., 1961). 

238 See, e.g., McGrain v. Daugherty, 273 U.S. 135, 174-75 (1927) (upholding investigatory 

powers of each house, including their power to compel testimony by subpoena); The 

Pacific Railroad Removal Cases, 115 U.S. 1, 14-15 (1885) (upholding Congress's power to 

charter railroad corporations); The Legal Tender Cases, 79 U.S. (12 Wall.) 457, 544, 553 

(1870) (upholding Congress's power under Necessary and Proper Clause to make treasury 

notes legal tender in payment of all debts and obligations).



1869 

resolution that obtains bicameral approval and the signature of the 

President. The "law" adopted authorizes the President to conclude an 

agreement that he has already or will subsequently negotiate.P 9 In 

this particular respect, Congress's approval of an international agreement 

is no different from its approval of any of the myriad agreements 

executive branch officials conclude in order to carry on the government's 

business. 240 Nor, despite Professor Tribe's heroic efforts to 

find a distinction, is it different from the long-settled congressional 

practice, sanctioned by the Supreme Court, of approving agreements 

negotiated by the President with Indian tribes. 241 Indeed, in Antoine 

239 For further discussion of the various forms congressional-executive agreements take, 

see Ackerman & Golove, supra note 5, at 804-05. 

240 See, e.g., 10 U.S.C.A. §§ 2302-2331 (West 1998) (providing ex ante authorization for 

military procurement contracts under strict congressional guidelines); 41 U.S.C. §§ 5-58 

(1994) (similar). 

241 Professor Tribe creates confusion by his conflation of two separate issues. The first is 

whether the term "laws" in the Necessary and Proper Clause includes a resolution approving 

an international agreement. The second is whether, notwithstanding an affirnative 

answer to the first question, Congress is prohibited from approving international agreements 

under that clause because of an "invisible radiation," Missouri v. Holland, 252 U.S. 

416, 434 (1920), from the Treaty Clause. At this juncture in his argument, Professor Tribe 

is challenging only the first, claiming that the linguistic terms of the Necessary and Proper 

Clause are not broad enough to encompass approving agreements. The accepted practice 

of Congress approving federal contracts and agreements with Indian tribes, however, conclusively 

resolves the point against his view. See supra note 240 and accompanying text; 

infra note 243 and accompanying text. Nevertheless, it is true that those same practices 

have no direct bearing on the second question, which, as I have repeatedly made clear, 

depends on other considerations. Thus, Professor Tribe's repeated emphasis on Congress's 

plenary power over Indian affairs is entirely beside the point. Congress likevise has plenary 

power over foreign affairs, but this fact is irrelevant to the question whether the term 

"laws" includes resolutions approving agreements. So too is the fact that international 

agreements are between sovereigns, whereas Indian agreements are between a sovereign 

(the United States) and its subjects (the Indian tribes). 

However, the latter distinction is relevant to the second question, whether congressional 

approval of Indian treaties demonstrates the nonexcusivity of the Treaty Clause. 

Ironically, Professor Tribe chides Professor Ackerman and me for having failed in our 

article to address the longstanding practice, going back to the 1870s, of congressional approval 

of Indian agreements. In 1871, Congress passed a statute declaring that Indian 

tribes are not independent nations with whom the United States may contract by treaty. 

See Act of Mar. 3, 1871, ch. 120, 16 Stat. 544, 566 (codified as amended at 25 U.S.C. § 71 

(1994)). This statute put an end to the practice of treatymaking with the Indians, and in its 

place, Congress began to approve agreements negotiated by executive branch officials, 

much as it approves congressional-executive agreements. Professor Tribe hypothesizes 

that we did not address this practice because "it does not enhance [our] depiction of 1945 

as a radical break with the past." Tribe, Taking Text, supra note 3, at 1262 n.138. This ad 

hominem is doubly ironic, first, because Professor Tribe himself, who accepts our view of 

the early history, see supra note 39, immediately proceeds to explain why this practice is 

entirely consistent with an exclusive reading of the Treaty Clause, and second, because our 

treatment of the subject ended up on the cutting room floor as a matter too obvious to 

require further elaboration. Alas.



[Vol. 73:1791 

v. Washington, 242 the Court explicitly held that congressional "ratification 

of an agreement between the Executive Branch and an Indian 

tribe is a '[Law] of the United States... made in Pursuance' of the 

Constitution. '' 243 There is, of course, an important difference: An international 

agreement creates obligations under international law. 

But that is relevant not to whether approving an agreement is a "law" 

but to whether the Treaty Clause implies that the international nexus 

In any case, the explanation is quite simple, although the history is complex and 

fascinating. As relations between the United States and the Indians developed, the early 

sui generis perception of the character of Indian sovereignty and of the relationship between 

the United States and the Indians came under increasing stress. The predominant 

view slowly shifted, culminating in a new conceptualization of the Indians as subjects and 

wards of the nation. At the same time, the House came increasingly to resent the regulation 

of Indian affairs through treaties, a process from which it was excluded, and began to 

protest against the application of a procedure designed principally for undertaking international 

obligations to domestic groups having, in their view, no international status. After a 

drawn out and bitter legislative struggle, the Senate capitulated and agreed to the 1871 

statute. See generally Antoine v. Washington, 420 U.S. 194, 201-03 (1975) (discussing conflict 

between House and Senate that resulted in 1871 statute). For excellent surveys of the 

history, see Felix S. Cohen's Handbook of Federal Indian Law 62-70, 105-07 (Rennard 

Strickland ed., 1982); Francis Paul Prucha, American Indian Treaties 289-310 (1994). Thus, 

far from supporting the nonexclusivity of the treaty procedure, congressional approval of 

Indian agreements simply marks the sharp historical differentiation between "genuine" international 

agreements and all other forms of legislation. Indeed, the point of the 1871 

statute was not to make the senatorial and congressional procedures interchangeable (i.e. 

nonexclusive), but to exclude the use of the senatorial procedure altogether, as inapplicable 

to Indian tribes. The statute thus raises an entirely different issue. No one in 1871 

imagined that the statute would have any bearing on the exclusivity of the treaty procedure 

within the "proper" field of its application to binding international commitments under the 

law of nations. Surprisingly, Professor Tribe intimates that in his view the 1871 statute 

might be unconstitutional because Congress cannot strip the President and the Senate of a 

constitutional authority. See Tribe, Taking Text, supra note 3, at 12& n.144. True, but if 

so, it would only be because the Framers in effect constitutionalized the status of Indian 

tribes in a way that makes application of the Treaty Clause to Indian agreements appropriate 

and renders impermissible a change in the conceptualization of the Indians to purely 

domestic subjects and wards of the state. Perhaps this is the better-though long unappreciated-view. 

But even if correct, this view says nothing about the constitutionality of 

the congressional-executive agreement. 

242 420 U.S. 194 (1975). 

243 Id. at 201 (quoting U.S. Const. art. VI, cl. 2). 

Professor Tribe chides Professor Ackerman and me for treating the distinction between 

making laws and approving agreements "dismissively." Tribe, Taking Text, supra 

note 3, at 1261. However, in the pages of our article to which he cites, we simply made the 

point that his position at that time undermined the very distinction he was trying to draw. 

He was then claiming that Congress could approve most international agreements but that 

some very important agreements had to be processed exclusively as treaties. In response, 

we noted that this position was inconsistent with the distinction he had drawn between 

laws and approving agreements, for if congressional resolutions approving less important 

agreements are "laws" for purposes of the Necessary and Proper Clause, then resolutions 

approving even important agreements must be "laws" as well. See Ackerman & Golove, 

supra note 5, at 919-22. In response, of course, Professor Tribe radically shifted his view; 

he now holds that Congress cannot approve any agreements.




can be tied only by the President and the Senate. Thus, we return full 

circle to our initial question. 

Professor Tribe's real objection, then, must not be as to form but 

substance. Even though congressional approval may take the form of 

a law, it is not, he thinks, a legislative act. But this reasoning too is 

obscure. Chadha, upon which he heavily relies, teaches that an act is 

legislative in character when it alters "the legal rights, duties, and relations 

of persons... outside the Legislative Branch." 244 These persons 

may include, moreover, "Executive Branch officials." 2 45 Clearly, authorizing 

the President to conclude an agreement alters his legal rights 

and, upon the contingency of his depositing an instrument of ratification, 

the legal rights and duties of those to whom the agreement, as 

supreme law of the land, will apply. It would seem, then, to pass the 

Chadha test with flying colors. 

Professor Tribe relies principally on the Framers' intent, citing 

statements by Hamilton and Madison he construes as denying the legislative 

character of approving agreements. This is odd given his adherence 

to an original meaning interpretive methodology, 46 and 

unhelpful since he seriously misinterprets their remarks. Odder still, 

however, is his lack of attention to the textual evidence. One has to 

suspect that he is dissatisfied with the results of textual exegesis-and 

for good reason. If we consider text alone, while there are arguments 

going both ways, the weight of the evidence tips decidedly in favor of 

the legislative side. The Treaty Clause, of course, appears in Article II 

and is framed as a power granted to the President. Taken alone, this 

would certainly suggest that the Framers viewed approving agreements, 

like negotiating them, as executive acts. But this is assuredly 

not the only evidence the Constitution yields. Thus, while the text 

places the treaty power among those granted to the President, it subjects 

the President's power to strict legislative oversight, requiring a 

special majority of the Senate to approve his agreements. Had the 

Framers believed that approving agreements was an executive act, 

surely they would not have subjected it to a supermajority legislative 

veto. This suggests that while negotiating agreements is an executive 

task, approving agreements is legislative in character. 247 This view is 

244 I.N.S. v. Chadha, 462 U.S. 919, 952 (1983). 

245 Id. 

246 Perhaps Professor Tribe believes recourse to historical evidence is appropriate in 

determining whether people at the Founding would have viewed approving an agreement 

as a legislative act. See supra notes 38-39 and accompanying text. 

247 Madison emphasized this point in arguing that the treaty power is legislative: "One 

circumstance indicating this, is the constitutional regulation under which the senate give 

their consent in the case of treaties. In all other cases, the consent of the body is expressed 

by a majority of voices." Helvidius No. 1, supra note 69, at 147-48. Here, the contrast with


1872 


[Vol. 73:1791 

confirmed, moreover, by two other provisions. First, the Compact 

Clause subjects state agreements and compacts to congressional approval. 

2 48 This both confirms that approving agreements is a legislative 

act and undermines any (already strained) inference from the 

Treaty Clause that it is somehow senatorial but not legislative in character. 

Second, the Supremacy Clause provides that the Constitution, 

the laws of the United States, and treaties made under its authority 

"shall be the supreme Law of the Land. 2 49 Since treaties are 

supreme law, it stands to reason that approving them is a legislative 

act. 

50 ° 

Indeed, it would be extraordinary if such a sweeping power to 

create supreme law of the land were a purely executive function. 

the Appointments Clause is of significance. In that clause, the Framers subjected the 

President's appointments only to a bare majority senatorial veto. This departure from 

their usual requirement of a two-thirds vote when one of the regular lawmaking organs 

does not participate tends to suggest the executive character of appointments in contrast to 

the legislative character of approving agreements. See supra notes 69-74 and accompanying 

text. Madison held just this view. See supra notes 167-81 and accompanying text; infra 

notes 250, 253, 256, and accompanying text. 


249 Id. art. VI, cI. 2 (emphasis added). 

250 Professor Tribe points to the Supremacy Clause as evidence that approving agreements 

is not legislative. He does so, however, only by lifting part of the clause out of 

context. Thus, he misleadingly quotes only the first part of the clause, which refers to "'the 

Laws of the United States which shall be made in Pursuance [of the Constitution]; and all 

Treaties made, or which shall be made, under the Authority of the United States,"' Tibe, 

Taking Text, supra note 3, at 1262 n.136 (quoting U.S. Const. art. VI, cl. 2), and argues that 

this language demonstrates that the Framers viewed lawmaking and treatymaking as different 

kinds of acts. He fails, however, to quote the critical portion of the clause that directly 

follows his quotation-both laws and treaties, it provides, "shall be the supreme Law of the 

Land." U.S. Const. art. VI, cl. 2. Of course, the Framers distinguished between laws and 

treaties to provide a special procedure for approving the latter. That is why there is a 

Treaty Clause. But both laws and treaties, the Supremacy Clause tells us, are the "Law" of 

the land. As Madison put it in arguing for the legislative character of treatymaking: 

[T]reaties, when formed according to the constitutional mode, are confessedly 

to have force and operation of laws, and are to be a rule for the courts in 

controversies between man and man, as much as any other laws. They are 

even emphatically declared by the constitution to be "the supreme law of the 

land." 

Helvidius No. 1, supra note 69, at 148. 

Professor Tribe attempts one other thrust at textual argument. He notes that Article I, 

Section 1 provides that all "legislative Powers herein granted" shall be vested in Congress. 

U.S. Const. art. I, § 1. Since Article II vests the treaty power in the President and the 

Senate, ipso facto, he concludes, it must not be a legislative power. Of course, Articles II 

and III likewise "vest" the executive and judicial powers in the President and the courts, 

respectively. Professor Tribe's notion that the three types of powers neatly divide into 

hermetically sealed boxes, then, is entirely inconsistent with the many powers that are 

overlapping and concurrent. See supra notes 117-23, 131-34, 194-209, and accompanying 

text; infra notes 398-403 and accompanying text. Perhaps no mechanical formalist argument 

has more fully merited Justice Holmes's famous quip: "The great ordinances of the 

Constitution do not establish and divide fields of black and white." Springer v. Philippine 

Islands, 277 U.S. 189, 209 (1928) (Holmes, J., dissenting).




1873 

When we turn to the historical materials, the evidence is equally 

strong that the Framers-or at least many among them-viewed approving 

agreements as legislative. It is true that a number of early 

theorists, including Locke and Montesquieu, had viewed it as an executive 

authority.2 5 ' But they wrote during the period when the modem 

treaty power was still in formation, and their views were largely an 

artifact of earlier conceptions of sovereignty as a personal attribute of 

the monarch.2 5 2 In any case, although there was a wide spectrum of 

views during the ratification period, key Framers clearly articulated 

the view that approving treaties is a legislative function. During the 

Convention, Madison, Wilson, and Mason all expressed this conviction,2 

3 and both Washington and Jefferson later pronounced the same 

position.2 54 In The Federalist, Hamilton too said approving treaties 

251 Locke viewed all powers relating to external affairs, including the treaty power, as 

belonging to the "federative" power, which was distinct from the executive power. Nevertheless, 

both powers, he thought, ought to be united in one officer. See John Locke, Second 

Treatise of Civil Government § 146 (Crawford B. Macpherson ed., Hackett 1980) 

(1690). Montesquieu placed all external powers in the executive. See Charles de Seconat, 

Baron de Montesquieu, The Spirit of the Laws, bk. XI, ch. 6 (Anne M. Cohler et al. eds. & 

trans., Cambridge Univ. Press 1989) (1748). 

252 See Peter Haggemnacher, Some Hints on the European Origins of Legislative Participation 

in the Treaty-making Function, in Parliamentary Participation in the Making and 

Operation of Treaties 19, 19-26 (Stefan A. Riesenfeld & Frederick M. Abbott eds., 1994). 

Professor Haggenmacher's excellent study of the early intellectual history explains that the 

use of treaties as a general instrument for conducting foreign affairs was still nascent at the 

time these authors wrote. Peace treaties were their almost exclusive form, and as such they 

were perceived as inextricably tied to the war power, which belonged to the sovereign 

(though here too the Framers departed from tradition). Also critical were changing conceptions 

of sovereignty, which evolved from a personal right inhering in certain persons to 

an attribute of nationhood residing in an absolute sovereign. Finally, as the implications of 

emerging notions of popular sovereignty became more clear, the attitudes of leading theorists 

underwent a corresponding shift. See infra notes 257-58 and accompanying text. 

253 See 1 Records of the Convention, supra note 187, at 65-66, 73-74 (remarks of James 

Wilson) (declaring treaty power to be legislative and reflecting even in 1787 close historical 

association between treaty power and war power in referring to former as power to make 

peace); 1 id. at 70 (remarks of James Madison) (same); 2 id. at 522-23, 530 (remarks of 

James Wilson) (same); 2 id. at 537 (remarks of George Mason) (same). Madison expressed 

the same view on numerous other occasions as well. See supra notes 167, 181, 250, and 

accompanying text; infra note 256. 

254 See Jefferson, supra note 47, § 52, at 298 ("Treaties are legislative acts. A treaty is 

the law of the land. It differs from other laws only as it must have the consent of a foreign 

nation, being but a contract with respect to that nation."); Washington, supra note 181, at 

378 (contrasting, in remarks to Senate committee, executive character of Senate's participation 

in appointments with acknowledgment that "[i]n treaties, the agency is perhaps as 

much of a legislative nature"). Professor Yoo concludes that the Framers believed 

treatymaking "to be a legislative function." John C. Yoo, The Continuation of Politics by 

Other Means: The Original Understanding of WVar Powers, 84 Cal. L Rev. 167,249 (1996); 

see also id. at 249 & nn.399-400 (construing John Jay's view as in accord). For the best 

accounts of the Framers' deliberations, see Arthur Bestor, Respective Roles of Senate and 

President in the Making and Abrogation of Treaties-The Original Intent of the Framers



was more legislative than executive in nature, 255 and Justice Story confirmed 

this view in his Commentaries, when he recalled the debates at 

the Founding. 2-6 Perhaps in this respect the Framers were influenced 

by their reading of Vattel, who argued in 1758 that the treaty power 

belongs to those in whom the sovereignty of the state ultimately resides: 

If absolute sovereignty is vested in the monarch, then the treaty 

power belongs solely to him, but if sovereignty resides in the people, 

then it is up to the fundamental law to determine which authority is 

capable of exercising it.257 As a consequence, in some states, he 

pointed out, the executive has "to take counsel of a senate or of the 

258 s 

representative body of the Nation. 

Professor Tribe takes no note of any of this history. Instead, he 

relies principally on a statement of Hamilton's in Federalist No. 75. In 

this quotation, Hamilton asserts that making treaties is not strictly a 

legislative or an executive power. From this, Professor Tibe infers 

that the legislative power does not include the power to approve international 

agreements, claiming that "it was clear to Hamilton, at least, 

that the making of international agreements was not a task for Congress." 

59 To reach this conclusion, however, he disregards Hamilton's 

express statements to the contrary and seriously mangles the quote. 

of the Constitution Historically Examined, 55 Wash. L. Rev. 1, 73-132 (1979); Jack N. 

Rakove, Solving a Constitutional Puzzle: The Treatymaking Clause as a Case Study, 1 

Persp. Am. Hist. 233, 236-50 (1984). This is not to deny that some contemporaries may 

have believed that agreement-making was executive in character, though Professor Tribe 

has not mustered any such evidence. At a minimum, nothing in the historical evidence 

renders the Necessary and Proper Clause an implausible textual grounding for Congress's 

power to approve agreements falling within its legislative authority. 

255 See The Federalist No. 75, supra note 187, at 450. In the heat of battle over 

Washington's unilateral declaration of neutrality between the British and the French, however, 

Hamilton, writing as Pacificus, switched views. See Pacificus No. 1, supra note 106, at 

33, 39, 42. This provoked Madison's vigorous denial in his Helvidius letters. See Helvidius 

No. 1, supra note 69, at 143-50. Madison reaffirmed his earlier view and, in his concluding 

riposte, charged Hamilton with having borrowed his doctrine, not from the Constitution, 

but from the "royal prerogatives in the British government." Id. at 150. Inexplicably, Professor 

Tribe relies on Hamilton's initial view in Federalist No. 75, rather than his later 

turnabout. See infra notes 259-60 and accompanying text; infra note 261. 

256 See 3 Story, supra note 202, § 1508, at 361 (discussing debates over character of 

treaty power); 3 id. § 1513, at 365-66 (concluding that it "partake[s] more of the legislative, 

than of the executive character"). 

257 See 3 Emmerich de Vattel, The Law of Nations or the Principles of Natural Law, bk. 

2, ch. 12, § 154, at 160 (Charles G. Fenwick trans., Oceana Publications 1964) (1758). 

258 Id. Vattel, with whom the Framers were apparently familiar, see United States Steel 

Corp. v. Multistate Tax Comm'n, 434 U.S. 452,462 n.12 (1978) (discussing Framers' familiarity 

with Vattel), was a populizer of the earlier work of Christian von Wolff. Some hints 

of Vattel's views can even be found in Grotius. For further discussion, see Haggenmacher, 

supra note 252, at 27-28. 




1875 

To correct Professor Tribe's error, it is necessary to quote Hamilton at 

some length: 

Though several writers on the subject of government place [the 

treaty] power in the class of executive authorities, yet this is evidently 

an arbitrary disposition; for if we attend carefully to its operation 

it will be found to partake more of the legislative than of the 

executive character, though it does not seem strictly to fall within 

the definition of either of them. The essence of the legislative authority 

is to enact laws, or, in other words, to prescribe rules for the 

regulation of the society; while the execution of the laws and the 

employment of the common strength, either for this purpose or for 

the common defense, seem to comprise all the functions of the executive 

magistrate. The power of making treaties is, plainly, neither 

the one nor the other.... Its objects are CONTRACTS with foreign 

nations which have the force of law, but derive it from the obligations 

of good faith.... The power in question seems therefore to 

form a distinct department, and to belong, properly, neither to the 

legislative nor to the executive. 60 

The key points are first that Hamilton affirms the treaty power as 

more legislative than executive in character, although, ironically, Professor 

Tribe later goes on to ignore Hamilton altogether and claim 

agreement-making as a solely executive function. 2 61 Second, 

Hamilton is discussing the treaty power as a whole when he says that 

it does not fall strictly into either camp. This does not mean, as Professor 

Tribe would suppose, that no part of it belongs either to the 

legislative or the executive branch, but that it ought properly to be 

divided between them. Thus, Hamilton's view quite nicely explains 

why the negotiating of an agreement ought to be given to the executive 

but its approval be reserved to the legislative branch. Indeed, he 

goes on in the next sentence to explain as much: 

The qualities elsewhere detailed as indispensable in the management 

of foreign negotiations point out the executive as the most fit 

agent in those transactions; while the vast importance of the trust 

and the operation of treaties as laws plead strongly for the participation 

of the whole or a portion of the legislative body in the office of 

making them. 262 

260 The Federalist No. 75, supra note 187, at 450-51. Professor Tribe conveniently omits 

the entire first sentence except for the last clause, thus beginning with "though it does not 

seem strictly to fall with the definition of either of them." See Tibe, Taking Text, supra 

note 3, at 1262. 

261 See Tribe, Taking Text, supra note 3, at 1265. As previously noted, Hamilton did in 

fact change his views subsequently, see supra note 256, but Professor Tribe relies on 

Hamilton's initial position in The Federalist, not his later turnabout. 

262 The Federalist No. 75, supra note 187, at 451 (emphasis added).



[Vol. 73:1791 

Hamilton then goes on to explain why requiring the President to 

obtain the approval of the House as well as the Senate would create 

special problems in light of the peculiar institutional characteristics of 

that part of the legislative branch. Thus, while Hamilton may have 

viewed "treatymaking" as not "strictly" a legislative task, approving 

agreements fell directly within the legislature's domain. 

Professor Tribe's reliance on Madison is equally puzzling. He 

quotes from Madison's speech in the House during the famous debate 

over the first Bank of the United States. In the course of defending a 

narrow reading of Congress's implied powers, Madison said: "Had 

the power of making treaties .. been omitted, however necessary it 

might have been, the defect could only have been lamented, or supplied 

by an amendment of the Constitution. 2 63 This Professor Tribe 

takes to prove his claim that approving agreements is not a legislative 

act. Presumably, he thinks (inexplicably) that Madison would have 

affirmed an unenumerated congressional power to approve agreements 

had he believed it to be legislative in character. Again, however, 

Professor Tribe is lifting comments out of context and thereby 

missing their clear import. Not only did Madison affirm his view of 

treatymaking as legislative both before and after these remarks, 2 64 but 

there is nothing in the quoted statement to the contrary. Indeed, 

Madison's comment says nothing whatsoever about the character of 

agreement-making as legislative or otherwise or even about Congress's 

power under the Necessary and Proper Clause to approve 

agreements incident to its enumerated foreign affairs authority. 

Immediately prior to the quoted remarks, Madison distinguished 

between two kinds of implied powers-those necessary and proper for 

executing an enumerated power and those necessary and proper for 

the Government or Union. The former were permissible under the 

Necessary and Proper Clause insofar as they were evidently and necessarily 

incident to an express power. The latter, however, were never 

permissible: "This constituted the peculiar nature of the Government; 

no power, therefore, not enumerated could be inferred from the general 

nature of Government. ' 2 65 He then offered the treaty power as an 

example. No matter how crucial it might be to the operation of the 

federal government, if it had not been expressly granted it could not 

be inferred from the general nature of the government. Although 

263 2 Annals, supra note 230, at 1900-01 (remarks of Rep. Madison). 

264 See supra notes 167, 181, 250, 253, 255, and accompanying text. 

265 2 Annals, supra note 230, at 1900 (remarks of Rep. Madison) (emphasis added).



187 

Madison's proposition is not categorically true, 2 6 it is still a sensible 

interpretation of the Constitution and one with which Chief Justice 

Marshall concurred in McCulloch. Marshall explicitly denied that a 

general congressional power to incorporate could be implied: 

Had it been intended to grant this power as one which should be 

distinct and independent, to be exercised in any case whatever, it 

would have found a place among the enumerated powers of the 

government. But being considered merely as a means, to be employed 

only for the purpose of carrying into execution the given 

powers, there could be no motive for particularly mentioning it.2 6 7 

The same analysis, consistent with Madison's remarks, applies to the 

power to approve agreements. Had the Framers meant to grant Congress 

a general power to approve any and every agreement, whatever 

its subject matter, they would likely have included such a power 

among the others expressly given. But the congressional-executive 

agreement does not depend on any such general power in Congress. 

Congress has the power to approve an agreement only as a means to 

execute its foreign affairs powers. As a means, "there could be," in 

Chief Justice Marshall's words, "no motive for particularly mentioning 

it."268 

C. Affirmative Structural Arguments 

I turn now to the structural arguments. Here, again, I divide 

them into affirmative arguments for exclusivity and negative arguments 

against Congress's power to approve agreements. In this section, 

I deal only with the former. Professor Tribe makes only one 

266 In Curtiss-Wright, the Court found the whole of the external affairs powers to be 

implicit in the nature of the federal government. See supra notes 113-15 and accompanying 

text; infra note 347. 

267 McCulloch v. Maryland, 17 U.S. (4 Wheat.) 316, 421-22 (1819). 

68 Id. at 422. To be sure, Madison's constricted interpretation of the Necessary and 

Proper Clause, rather than the explicit reference to the treaty power upon which Professor 

Tribe relies, might well be inconsistent with the congressional-executive agreement. 

Madison argued that only those powers that were evidently and necessarily incident to an 

enumerated power were to be implied. Even this category had to be narrowed to exclude 

any powers that were inherently important powers in themselves. In the latter category, he 

placed the naturalization power, which he argued would have been too important to be 

implied under the Necessary and Proper Clause had it been omitted from the text, even if 

it were necessarily incident to an express power. See 2 Annals, supra note 230, at 1899- 

1901 (remarks of Rep. Madison). These arguments, however, are irrelevant for us. 

McCulloch swept them into the dustbin of history where they have remained ever since. 

Moreover, even Madison's specific claim that the naturalization power was too important 

to be implied has long been discredited. Although the text in fact provides for naturalization, 

it says nothing about the immigration power-the power to admit, exclude, and expel 

aliens. Yet, the Court long ago had no trouble finding plenary power in Congress over this 

equally important and closely related subject. See supra note 115 and accompanying text.



[Vol. 73:1791 

affirmative structural argument. Although he never articulates the 

point clearly, he invokes the federalism argument that I identified at 

the outset as one of the two main arguments in favor of the exclusive 

view. But as in the case of his expressio unius argument, he seeks to 

endow it with far more weight than it can sustain. 

1. Exclusivity, Basic Framework Provisions, and Federalism 

Professor Tribe claims that certain kinds of constitutional provisions 

ought by their nature to be given an exclusive sense. He identifies 

these as "provisions that establish the basic framework of our 

system of governance" 269 and, in a slightly different formulation, "provisions 

that specify how elements of the supreme law of the land are 

to be adopted." 270 These apparently include the procedures for proposing 

and ratifying constitutional amendments, for adopting laws, 

and for making treaties. Of course, there is no axiomatic link connecting 

these sorts of provisions to the conclusion that they ought to be 

exclusive. What underwrites Professor Tribe's view is structural- 

"the overarching concern with state sovereignty" that animated the 

Framers' design. 271 Exclusivity is necessary to uphold "the careful efforts 

of the Framers to establish an elaborate scheme of checks and 

balances and a delicate division of lawmaking power. '272 

But why does this conclusion follow from the premise To be 

sure, from an originalist perspective, care should be taken to avoid 

substantially undermining the Framers' delicate division of powers between 

the federal and state governments. However, construing a basic 

framework provision as nonexclusive will not necessarily create 

this evil. In some instances, that will be precisely what the Framers 

had in mind, in which case Professor Tribe's principle would obviously 

have no application. In others, their thinking may be less clear, but 

that is no reason to leap to an exclusive construction. Exclusivity will 

not necessarily best serve the interests of the states. Even when it 

does, moreover, state interests cannot be the only consideration that 

weighs in the constitutional balance, and the imperative to preserve 

federalism values may be outweighed by competing concerns. In each 

case, the answer will depend on the particular provision being 

construed. 


270 Id. at 1244. 

271 Id. 

272 Id. at 1241. For discussion of this argument not as a federalism principle but as an 

interpretation of the expressio unius maxim, see supra notes 124-38 and accompanying 

text.




It is clear that Professor Tribe assumes that the Framers designed 

the Treaty Clause to provide special protection for state interests and 

that the interchangeability doctrine undermines their scheme. This is 

indeed one of the two principal arguments in favor of exclusivity, but 

Professor Tribe never goes beyond mere assertion. 273 As I pointed 

out at the outset, however, the text-in contrast to the historical evidence 

274 -does not yield a single plausible inference, and there are 

persuasive textual and structural arguments to counter the states' 

rights interpretation of the Framers' intent. 275 Professor Tribe simply 

never confronts the crucial interpretive difficulties posed by these conflicting 

plausible inferences or tells us why his construction is any 

more persuasive than the alternative view. Indeed, it may well be his 

recognition of the equivocal nature of the textual evidence that leads 

him to prefer a sweeping presumption of exclusivity. He would simply 

require us to presume without analysis that basic framework provisions 

protect federalism interests that would necessarily be undermined 

by a nonexclusive reading. But he never says why we should 

undertake constitutional analysis on the basis of rigid presumptionspresumptions, 

indeed, which he spins entirely out of whole cloth. 

Furthermore, his distinction between basic framework provisions 

specifying how elements of the law are to be adopted and all others is 

not well drawn and when removed threatens to engulf his argument in 

the same morass that mires his expressio wuhs argument.2 76 All sorts 

of constitutional provisions potentially reflect a delicate division of authority 

between the federal and state governments. There seems to be 

no more reason to treat as categorically exclusive those on Professor 

Tribe's narrow list than any of the others granting constitutional powers. 

Yet, not all constitutional grants are exclusive. Thus, for example, 

when the Framers designed the electoral college system, they 

made important choices about how much to favor the interests of the 

whole over the interests of individual states in the office of the 

273 An assertion he repeats on many occasions. See, e.g., Tribe, Taking Text, supra note 

3, at 1228, 1241, 1248, 1282, 1292. In defense of his claim, he only points out that the 

Framers assigned the power of advice and consent to the Senate, in which the states are 

represented equally. See id. at 1241. But Professor Tribe fails to notice that the 

congressional-executive agreement also requires approval by the Senate; it just includes 

the House as well. What is crucially at stake, then, is not the involvement of the Senate, 

but the two-thirds voting requirement. 

274 See Ackerman & Golove, supra note 5, at 808-13 (concluding early understandings 

strongly support exclusivity). 

275 See supra notes 69-75, 186, and accompanying text; infra note 378. The two main 

competing conceptions are the states' rights view and the claim that the Treaty Clause just 

gives the President an extraordinary option to be used when he believes that considerations 

of secrecy, dispatch, or long-term perspective counsel exclusion of the House. 




[Vol. 73:1791 

President. This too represented a delicate division of power between 

the federal and state governments. As we have seen, the Court has 

nevertheless upheld the independent agencies, as it has countless 

other concurrent exercises of constitutional power. 27 7 Unless we wish 

to revisit these decisions, it simply cannot be the case that from the 

fact that a grant of constitutional power reflects a particular division 

of authority between the federal and state governments, it follows that 

it ought to be exclusive. 

The Court's approach to problems of exclusivity vis-a-vis the 

states provides an important illustration of this point. The provisions 

of Article I, Section 10 explicitly deny the states certain powers that 

Article I, Section 8 vests in Congress. 278 These exclusions surely reflect 

a delicate division of lawmaking powers between the federal and 

state governments and hence have strong claim, a la Tribe, to be read 

as the exclusive limitations on state authority. All the more so, then, 

because of the expressio unius inference which Professor Tribe finds 

so compelling in other contexts and because the very frame of the 

Constitution is of a limited cession of authority from the sovereign 

states to the federal government. Nevertheless, from the beginning, 

the Court has held that some of Congress's powers are exclusive of 

the states even though not expressly denied to them in Article I, Section 

10. Although the Court no doubt begins with a presumption of 

nonexclusivity, the ultimate decision will depend upon whether or not 

the best inference from context suggests that the Framers wished to 

assign the power exclusively to Congress. 279 Thus, contrary to Professor 

Tribe's claim, there is no a priori answer to these kinds of interpretive 

problems. The mere fact that a particular construction might alter 

the division of authority does not resolve the interpretive difficulties. 

Professor Tribe's distinction between basic framework provisions 

and all others, moreover, is not well drawn in another sense. If the 

underlying concern is about protecting the interests of the states from 

federal encroachment, it is unclear why his principle ought to be narrowly 

limited to questions of exclusivity. Why not, for example, insist 

277 See supra note 117 and accompanying text. Similar remarks are apt for the courts. 

Still, in a wide range of fields, the Court has repeatedly upheld Congress's power to create 

alternative adjudicatory tribunals outside of Article III. See supra note 118 and accompanying 

text. Even the choice of which powers to grant to Congress in Article I and which to 

the President in Article II reflect choices relevant to the division of federal and state 

power. Nevertheless, as we have also seen, many of these have been deemed nonexclusive. 

See supra notes 117-23, 131-34, 194-209, and accompanying text; see also infra notes 398- 

403 and accompanying text (discussing nonexclusivity of Supreme Court Original Jurisdiction 

Clause). 

278 See supra note 115. 

279 See id.


Dezember 1998] 


upon a general rule requiring strict construction of any powers 

granted to the federal government, such as, say, the commerce power 

That provision surely reflects a delicate division of authority between 

the state and federal governments and thus in Professor Tribe's constitutional 

universe-where the Framers' "overarching concern with 

state sovereignty" 0 is controlling-ought to be given a narrow construction 

in favor of the states. But as Professor Tribe's expansive 

reading of the Commerce Clause suggests, this is surely not a direction 

in which he would wish to go. 81 

Finally, even if we confine ourselves to Professor Tribe's narrow 

notion of basic lawmaking provisions, he moves too quickly to his 

strict rule of exclusivity. Even assuming arguendo the exclusivity of 

Article V's amendment procedures and of Article I's lawmaking provisions,282 

Professor Tribe neglects the text's special provisions for 

making maritime law. Recall that the admiralty power is granted explicitly 

to the courts.3 Surely, this is a provision specifying "how elements 

of the supreme law of the land are to be adopted.' ' 2 84 And 

certainly the design of the courts (whose appointment is vested in the 

President and the Senate) reflects a delicate division of power between 

the federal and state governments. According to Professor 

Tribe's principle, this suggests that Congress's (implied) plenary 

power over admiralty, long accepted by the Court and previously affirmed 

by Professor Tribe himself, 2 85 should be revised. 

When we attend closely to Professor Tribe's proposed rule of construction, 

it becomes evident that it is as arbitrary as the strict expressio 

unius interpretive regime he urges and as inconsistent with our 

constitutional tradition. Constitutional interpretation is more 

280 Tribe, Taking Text, supra 3, at 1244. 

281 See id. at 1295-96 (endorsing broad reading of Congress's powers under Commerce 

Clause). For further discussion, see infra notes 421-23 and accompanying text. Perhaps, 

though, this is where the Court is moving. See Printz v. United States, 117 S. Ct. 2365,2379 

(1997) (holding Commerce Clause does not authorize Congress to commandeer state executive 

officials to carry out congressional policy in the Brady Handgun Violence Prevention 

Act); Seminole Tribe of Florida v. Florida, 517 U.S. 44,47 (1996) (holding Congress lacked 

authority under Indian Commerce Clause to abrogate state's Eleventh Amendment immunity 

in challenge to Indian Gaming Regulatory Act); United States v. Lopez, 514 U.S. 549, 

561 (1995) (holding Gun-Free School Zones Act exceeded Congress's authority under 

Commerce Clause). 

282 We have already considered the complications for his view of exclusive Article I 

lawmaking posed by agency rulemaking, the common law, and presidential legislating in 

foreign affairs. See supra notes 118, 129-30, and accompanying text. I leave to Professors 

Ackerman and Amar arguments justifying a nonexclusive reading of Article V. For relevant 

sources, see supra note 17. 

283 See supra notes 131-34 and accompanying text. 





[Vol. 73:1791 

nuanced than such categorical rules allow. We cannot escape the necessity 

of giving careful attention to context and the full range of relevant 

considerations by opting for simple formalist interpretive 

solutions like those he now professes to accept. 

D. Negative Structural Arguments 

Professor Tribe presses two negative structural arguments to undermine 

the textual bases for the congressional-executive agreement. 

The first invokes the danger to the President's primacy over foreign 

affairs posed by the congressional-executive agreement. Here, Professor 

Tribe relies upon the Veto Override Clause to suggest fatal structural 

defects in the Necessary and Proper Clause argument. His 

second argument is an effort to shield the exclusivity of the Treaty 

Clause from the apparent dilemma posed by the unilateral executive 

agreement. I address each in turn. 

1. Fears About the President's Primacy in Foreign Affairs: 

The Veto Override Clause and the Case of the Hyperactive 

Foreign Affairs Congress (Rarely Seen in These Parts) 

We turn now to the argument that Professor Tribe appears to 

view as his ace-in-the-hole-the possibility that accepting the 

congressional-executive agreement somehow entails that an aggressive 

Congress could conclude an international agreement on its own, 

even over the fierce objections of a sitting President. Professor Tribe 

is understandably proud of this point, since in the many momentous 

Treaty Clause debates over the course of our history, no one has ever 

made it before. Yet, on reflection, he might have considered whether 

this oversight reflects not so much upon the intelligence of his forebears 

than on the implicit historical judgment that the argument is 

simply without merit. 

Professor Tribe's argument is simple: Imagine a Congress bent 

on having its way in all matters and wishing to force a bill down the 

President's throat. Under the Veto Override Clause, all it need do, 

should he refuse to sign, is to repass the bill by a two-thirds vote in 

each house. 286 Now imagine the same Congress equally insistent 

286 There are really two veto override clauses, which are the same in all relevant respects 

except that the first applies to bills, while the second applies to orders, resolutions, and 

votes. See U.S. Const. art. I, § 7, cls. 2, 3. I refer to them collectively hereafter as "the 

Veto Override Clause." Congressional-executive agreements are ordinarily, but not exclusively, 

approved by joint resolution. See Ackerman & Golove, supra note 5, at 804-05 & 

805 nn.9-10 (citing examples). The second veto override clause provides, "[elvery Order, 

Resolution, or Vote to which the Concurrence of the Senate and House of Representatives 

may be necessary... shall be presented to the President," but if vetoed by him, shall only



AGAINST FREE-FORM FORM£ALISM 

about its views in foreign affairs and wishing to force a particular international 

agreement, perhaps one it itself negotiated or received 

from some foreign quarter, down that same President's throat. What 

is to prevent it from approving a resolution directing the President to 

ratify the agreement, "presenting" him with the resolution, and, upon 

his veto, repassing it by a two-thirds vote in each house If Congress 

has the power under the Necessary and Proper Clause to authorize 

the President to conclude an international agreement he has negotiated, 

then, in Professor Tribe's view, under "a coherent Article I- 

based view," 287 it ought to have the power to direct him to ratify any 

agreement, regardless of his role in its promulgation. Either way, it 

might find concluding the agreement necessary and proper for carrying 

into execution its enumerated powers. And if it can muster the 

necessary votes, it ought, under the Veto Override Clause, to have the 

power to bypass the President entirely. "The Constitution's text is unequivocal," 

in Professor Tribe's view, in applying the override power 

to "'[elvery Order, Resolution, or Vote, to Which the Concurrence of 

the Senate and House of Representatives may be necessary.'M' s 

Never mind the extreme remoteness of the hypothetical.2 9 Because 

the theoretical possibility thus presented is "dramatically at odds with 

the well-accepted principle that the President is the primary representative 

of the nation in foreign affairs," 290 Professor Tribe believes that 

the only viable solution is to rule out the congressional-executive 

agreement altogether. Case closed! 291 

take effect if "repassed by two thirds of the Senate and House." U.S. Const. art. 1, § 7, ci. 

3. 


8 Id. at 1252 (quoting U.S. Const. art. I, § 7, cl. 3). 

289 So far as I am aware, Congress has never attempted anything of the sort. But see 

infra note 291 (describing more assertive congressional actions in recent years). 


291 Even though Congress has never sought to negotiate an agreement on its owvn, Professor 

Tribe may be unaware of how close it has come towards this line in recent years. 

According to Professor Henkin, the so-called independent agencies have increasingly acted 

unilaterally on the international level, even at times negotiating agreements pursuant to 

congressional authority "without the participation, scrutiny or even awareness of the 

President or the Department of State." Henkin, supra note 9. at 129-30. This observation 

prompts Professor Henkin to query whether "the President is entitled to insist that he... 

be fully informed, and participate in any negotiations at least as an observer." Id. at 130 

(emphasis added). Surely, Professor Tribe would object to this practice, as he does to the 

congressional-executive agreement. Perhaps, it is unconstitutional on other grounds. 

However the issue is resolved, the simple fact that it describes the actual operation of our 

government tends strongly to undermine his claim that independent congressional 

agreement-making is entirely foreign to our practices. Cf. Trade Expansion Act of 1962, 

Pub. L. No. 87-794, § 243, 76 Stat. 872, 878 (requiring President to include members of 

Congress on delegations negotiating international trade agreements and thus arguably intruding 

on President's control over negotiating process), repealed by Trade Act of 1974,



[Vol. 73:1791 

Despite Professor Tribe's certainties about the clarity of the text, 

the application of Article I, Section 7's presentment and veto override 

provisions to the congressional-executive agreement is complex, providing 

some intriguing material for a law school exam in advanced 

constitutional law. As we shall see, closely related questions about the 

war powers have been the subject of long historical speculation among 

scholars, but remain unresolved since they have never arisen in practice. 

On this point, the opposite of Chief Justice Marshall's remark 

about the power of judicial review holds: Although the question 

whether Congress can override the President's veto in this area is of 

trivial practical importance to the United States, unhappily it is of an 

intricacy precisely disproportionate to its interest. 292 Happily, however, 

we need not attempt a resolution here, for, as we shall see, Professor 

Tribe's argument fails on other grounds in any case. 

Professor Tribe does heavily emphasize the veto override provisions. 

On close examination, however, it is clear that the core of his 

argument lies elsewhere and depends upon two entirely separate propositions. 

He believes that independent congressional agreementmaking 

is fatally inconsistent with the President's primacy over foreign 

affairs. 293 But he also thinks that from a textual perspective one 

cannot plausibly embrace both the congressional-executive agreement 

and any limits on Congress's independent agreement-making powers. 

Hence, the unconstitutionality of the modem procedure. 2 94 Both of 

these claims are essential to his argument, but neither is convincing. 

Pub. L. No. 93-618, §602(d), 88 Stat. 1978, 2072 (1975); Henkin, supra note 9, at 395 n.88 

(describing same). 

292 See Marbury v. Madison, 5 U.S. (1 Cranch) 137, 176 (1803) ("The question, whether 

an act, repugnant to the constitution, can become the law of the land, is a question deeply 

interesting to the United States; but, happily, not of an intricacy proportioned to its interest."). 

Professor Tribe considers and rejects an argument suggested, though not endorsed, 

by Professor Henkin, that would prohibit Congress from overriding a veto in those instances 

when its legislation is designed to implement a presidential power under the Necessary 

and Proper Clause, rather than a power of its own. See Tribe, Taking Text, supra note 

3, at 1256 n.117 (citing Henkin, supra note 28, at 915 n.26). I would reject this argument as 

well and would also reject Professor Tribe's implication that agreement-making is, in this 

sense, an executive power. For a different argument for avoiding the override power, 

which Professor Tribe considers and rejects, see id. at 1254 n.112 (rejecting, correctly, hypothetical 

argument that because treaty procedure is always available as an alternative, use 

of congressional-executive agreement is not a vote to which concurrence of both houses is 

"necessary," as required by Veto Override Clause). For other arguably more persuasive 

limitations on the veto override power, see infra notes 304, 308, and accompanying text. 

293 By "independent congressional agreement-making," I mean any effort by Congress 

to negotiate or ratify an international agreement either on its own or by directing the 

President to implement its policy. 

294 The veto override power enters into this argument only indirectly. Because the 

President could veto any congressional resolution directing him to ratify a congressionalexecutive 

agreement, the constitutional problem Professor Tribe identifies would ordina-




The first proposition-the inconsistency of independent congressional 

agreement-making with presidential primacy-may well be defensible 

on historical grounds. But by his own methodological fiat, 

Professor Tribe is bound to eschew history in favor of text and structure. 

It is here that he runs into trouble, for they both strongly suggest 

the contrary view. 295 Begin with the text. It is not entirely clear why 

Professor Tribe believes that the idea that Congress can bind the 

United States to an agreement by overriding the President's objections 

poses an insurmountable textual problem. 2 96 In support of his 

claim, he rests only on a citation to Justice Sutherland's opinion in 

United States v. Curtiss-Wright Export Corp. 2 97 As an adherent of the 

original meaning school, however, Professor Tribe would seem 

rily not arise. It can occur only if Congress overrides the President's veto. But, as we shall 

see, in objecting to such a resolution, the President would be most unlikely to point to the 

Veto Override Clause as the problem; his objection would be based on substantive constitutional 

law. When Congress approves a congressional-executive agreement, it simply 

passes a resolution authorizing the President to conclude the agreement on the international 

level As a result, the President would be most unlikely to veto the resolution of 

approval. Not requiring him to do anything in particular, he would have little reason to 

oppose it. Were he nevertheless to exercise the veto-and were Congress even more inexplicably 

to override-there would still be no real constitutional issue. The override would 

presumably be valid, but the resolution would only amount to an authorization that the 

President could freely disregard. On the other hand, were Congress to pass a resolution 

directing him to ratify an agreement, the President would be most likely to veto it (unless 

he favored the agreement, in which case he might just register his objections). And if 

Congress overrode his veto, he would most likely be unwlling to abide by its command. 

This result, however, would have nothing whatever to do with the Veto Override Clause, 

but rather with the President's view that the resolution, by directing him to ratify an agreement, 

had unconstitutionally invaded his exclusive sphere. The override would not itself 

be unconstitutional; it would be the bill's substance that would provoke objection. 

295 This may lead one to wonder why Professor Tibe has not broadened the focus of his 

textual attack to include not only the congressional-executive agreement but the 

President's non-textually-based sole organ power as well. While both practices find their 

historical roots in practice, the latter, as we shall see, provides a far more congenial context 

for a convincing textualist challenge. See infra notes 309-18 and accompanying text. 

296 Although he does not make the argument, perhaps Professor Tribe's sense of symmetry 

is offended by the idea that Congress can make agreements on its own but that the 

Senate, under the Treaty Clause, cannot. Yet, this seems perfectly consistent with the 

Framers' larger design: Congress can make an agreement over the President's veto, because 

when two-thirds of both houses are in favor, there is a strong enough national consensus 

to override the President's objections. However, the Senate acting alone could 

never reliably represent such a strong national consensus. 

2M 299 U.S. 304 (1936). Justice Sutherland's opinion has been roundly criticized for, 

among other things, its expansive dicta on the President's role in foreign affairs. See, e.g., 

Harold Hongju Koh, The National Security Constitution: Sharing Power After the Iran- 

Contra Affair 93-94 (1990) (criticizing opinion's key language as historically inaccurate and 

ambiguous dicta); Charles A. Lofgren, United States v. Curtiss-Wright Export Corporation: 

An Historical Reassessment, 83 Yale L.J. 1 (1973) (concluding Justice Sutherland provided 

no constitutional or historical ground for his broad view of broad, presidential foreign affairs). 

In favor of his propresidential bias, Professor Tribe, like Justice Sutherland, points 

to the President's superior access to, and ability to protect, confidential information neces-



[Vol. 73:1791 

obliged to point to convincing language in the text to justify his sweeping 

pro-Executive bias. Yet, it is notoriously difficult to find the 

needed support in the few scrawny provisions of Article II. And it 

becomes downright confounding when we compare those provisions 

to the rather hefty foreign affairs powers assigned to Congress in Article 

1.298 This is hardly fertile ground for an original meaning textualist 

argument. Indeed, the history of the constitutional law of foreign relations 

has been largely a search for nontextual justifications for the 

President's broad foreign affairs powers. Textualists have had to resort 

to highly attenuated extrapolations from modest textual provisions, 

and even then they have come up short. 2 99 Of course, this is 

true not only of the President's foreign affairs powers. As Professor 

sary to international agreement-making. See Tribe, Taking Text, supra note 3, at 1254-55 

(quoting with approval Curtiss-Wright, 299 U.S. at 320, on this point). 

298 Compare the President's power as Commander-in-Chief, see U.S. Const. art. II, § 2, 

cl. 1, to appoint and receive ambassadors and public ministers, see id. art. II, § 2, cl. 2, id. 

art. II, § 3, to make treaties, see id. art. II, § 2, cl. 2, and to take care that the laws are 

faithfully executed, see id. art. II, § 3, with Congress's power to declare war, see id. art. I, 

§ 8, cl. 11, to grant letters of marque and reprisal, see id., to make rules for captures on 

land and water, see id., to regulate foreign commerce, see id. art. I, § 8, cl. 3, to provide for 

the common defense of the United States, id. art. I, § 8, cl. 11, to raise and support armies, 

see id. art. I, § 8, cl. 12, to provide and maintain a navy, see id. art. I, § 8, cl. 13, to make 

rules for the government and regulation of the land and naval forces, see id. art. I, § 8, cl. 

14, to provide for organizing, arming, disciplining, and calling forth the militia, see id. art. I, 

§ 8, cls. 15, 16, to define and punish offenses against the law of nations and on the high 

seas, see id. art. I, § 8, ci. 10, to establish a uniform rule of naturalization, see id. art. I, § 8, 

cl. 4, to establish offices (including those pertinent to foreign affairs), see id. art. II, § 2, cl. 

2, and to make all laws necessary and proper for carrying into execution all of Congress's 

enumerated powers and the powers conferred on any other branch of the government, see 

id. art. I, § 8, cl. 18. In his Helvidius letters, Madison used this comparison to demonstrate 

that the conduct of foreign affairs is principally a legislative function given to the Congress. 

See Helvidius No. 1, supra note 69, at 145-48. 

299 See Henkin, supra note 9, at 31-36 (noting limited textual support for breadth of 

foreign affairs powers exercised by President). As Professor Henkin puts it: 

A stranger reading the Constitution would get little inkling of such large Presidential 

authority, and the general reader might comb the Constitution yet find 

little to support the legitimacy of large Presidential claims. The powers explicitly 

vested in him are few and appear modest, far fewer and more modest than 

those bestowed upon Congress. What the Constitution says and does not say, 

then, can not have determined what the President can and can not do. The 

structure of the federal government, the facts of national life, the realities and 

exigencies of international relations.. ., and the practices of diplomacy, have 

afforded Presidents unique temptations and unique opportunities to acquire 

unique and ever larger powers. 

Id. at 31. The desire to find a home in the text for the President's broad powers has also 

led some to repair to the Hamiltonian reading of the Executive Power Clause despite the 

linguistic leap of faith it demands and the ominous possibilities it opens. See id. at 40 

(discussing debate over Hamiltonian reading); see also supra notes 106-09 and accompanying 

text (discussing difficulty with this argument); infra note 342 and accompanying text 

(same).



AGAINST FREE-FORM FORA4LISM 

Tribe once eloquently put it: "To be xeminded that it was not meant 

to be so-that the Framers envisioned a vastly more modest chief 

magistrate-is only to recall that, had the blueprint been incapable of 

expanding beyond the Framers' designs, the Nation could not have 

persisted through two centuries of turmoil." 300 

In any case, the President's predominant role notwithstanding, 

the conundrum Professor Tribe identifies is built into the structure of 

the Constitution, and the larger difficulties are little affected by the 

case of the congressional-executive agreement. Congress has very 

wide, some say plenary, authority over foreign affairs. 301 Consider its 

power to regulate foreign commerce, to raise and support armies, and 

to define offenses against the law of nations. Yet, no one doubts that 

in exercising these extremely important foreign policy powers it may 

regulate even over the objections of the President. That is a necessary 

implication of the Veto Override Clause. Congress can even mimic 

through legislation most, if not all, of the obligations the country has 

undertaken through treaties. 30 2 To be sure, there would still be a difference; 

we would not be bound under international law to retain that 

legislation in place. The point, however, remains the same: Congress 

would be taking the lead in devising our foreign policy and could override 

any presidential opposition. Similarly, Congress can, and has, 

passed legislation inconsistent with our existing treaty obligations and 

thereby provoked our treaty partners (or the President anticipatorily) 

to terminate an agreement. 30 3 Here, too, it can, though it rarely has, 

acted against the President's wishes. 

Even waiving these points, the most damning textual rejoinder to 

Professor Tribe's view is found in Congress's most momentous foreign 

affairs prerogative-the power to declare war. Under a straightforward 

reading of Article I, Section 7, we are forced to concede that 

two-thirds of both houses can force the country into war even over the 

most vehement opposition of the President, notwithstanding his status 

as Commander-in-Chief. This certainly suggests that Professor Tribe's 

300 Tribe, Constitutional Law, supra note 11, § 4-1, at 209. Professor Tribe then diewed 

the textual grants of foreign affairs powers in Article If as "textual manifestations of the 

inherent presidential power" to administer the foreign relations of the nation. Id. § 44, at 

220. 

301 See Henkin, supra note 28, at 920-30 (arguing for plenary congressional power); see 

also Henkin, supra note 9, at 71-72 (same); supra notes 28, 48-52, 298. and accompanying 

text (discussing support for this view). 

302 See Henkin, supra note 9, at 71-72 ("[T]he Foreign Affairs Power would support 

legislation on any matter so related to foreign affairs that the United States might deal with 

it by treaty. .."); Henkin, supra note 28, at 920-30 (arguing same). 

303 For further discussion, see Corwin, The President, supra note 119, at 220-21 & 478 

n.58; Jesse S. Reeves, The Jones Act and the Denunciation of Treaties, 15 Am. J. Int'l L 33 

(1921).



[Vol. 73:1791 

simplistic affirmation of the President's primacy in foreign affairs is 

out of step with the constitutional structure. It perhaps adds insult to 

injury to note that most of the historical speculation has been over 

whether Congress's control over war and peace is so complete that the 

President's veto power ought somehow to be deemed inapplicable to 

a declaration of war. 304 Presidents, in any case, have at times been 

careful to delay convening Congress or releasing inflammatory information 

in order to avoid inciting Congress and provoking it to war. 30 5 

Still, even leaving this additional complication aside, Professor Tribe's 

architectural worries about Congress somehow binding the country to 

an agreement without the President's sanction pales in comparison to 

this vast and portentous authority in the Congress alone. If, despite 

the President's predominance in foreign affairs, Congress can make 

the awesome decision to thrust the country into war unilaterally, why 

should the President's supposed primacy rule out congressional 

agreement-making 

The answer seems all the more clear when one considers that the 

Framers explicitly allowed Congress just such a power. Under the 

304 Some have claimed that a declaration of war ought not be subject to the veto, see, 

e.g., 28 Cong. Rec. 2107 (1896) (remarks of Sen. Morgan); most have disagreed, see, e.g., 

Berdahl, supra note 118, at 79-80, 95-96 (arguing that declaration of war is, like any other 

bill or resolution, subject to veto); Clinton Rossiter, The Supreme Court and the Commander 

in Chief 66 n.1 (Richard P. Longaker ed., rev. ed. 1976) (same); James Schouler, 

Constitutional Studies, State and Federal 137 (New York, Dodd, Mead & Co. 1897) 

(same); Simeon E. Baldwin, The Share of the President of the United States in a Declaration 

of War, 12 Am. J. Int'l L. 1 (1918) (same). Grover Cleveland was reportedly prepared 

to veto a possible declaration of war against Spain. See Rossiter, supra, at 66 n.1. More 

recently, the decision in I.N.S. v. Chadha, 462 U.S. 919 (1983), and its application to the 

War Powers Resolution, 50 U.S.C. §§ 1541-1548 (1994) (giving Congress legislative veto 

over use of troops in hostilities), stirred debate over the issue. Some have argued that 

Section 1545(c)'s legislative veto provision is not unconstitutional because the President 

has no authority to veto a declaration of war and thus Congress need not "present" a 

declaration for his consideration. See, e.g., Stephen L. Carter, The Constitutionality of the 

War Powers Resolution, 70 Va. L. Rev. 101, 129-32 (1984); Leonard G. Ratner, The Coordinated 

Warmaking Power-Legislative, Executive, and Judicial Roles, 44 S. Cal. L. Rev. 

461, 478-80, 489 (1971); War Powers and the Responsibility of Congress, 82 Am. Soc'y Int'l 

L. 1, 3-5 (1990) (remarks of Sen. Brock Adams and Prof. Louis Henkin). Still more recently, 

a scholarly debate has opened on the issue. Professor William fteanor, an excellent 

legal historian, claims that the original intent of the Framers was that a declaration was not 

subject to the veto, although admittedly I find the sources upon which he relies less convincing 

than he. See William Michael Treanor, Fame, the Founding, and the Power to 

Declare War, 82 Cornell L. Rev. 695,724-29 (1997). Compare Philip Bobbitt, War Powers: 

An Essay on John Hart Ely's War and Responsibility: Constitutional Lessons of Vietnam 

and Its Aftermath, 92 Mich. L. Rev. 1364, 1385 n.69 (1994) (reaching same conclusion on 

textual grounds), with John Hart Ely, War and Responsibility: Constitutional Lessons of 

Vietnam and Its Aftermath 231 n.21 (1993) (taking opposing view). 

305 For an account of President Jefferson's efforts to avoid war with Great Britain in 

1807, President Grant's in 1869, and Presidents Cleveland and McKinley's with respect to 

Spain in the later years of the nineteenth century, see Berdahl, supra note 118, at 84-92.




Compact Clause, states may enter into agreements or compacts with 

foreign states with Congress's consent. 306 If we accept Professor 

Tribe's understanding of the veto override power (as we should), it is 

clear that Congress can approve a state negotiated agreement even 

over the President's veto. Apparently, then, the idea was not so foreign 

to the Framers' scheme as Professor Tribe supposes. The Framers 

explicitly provided for just such a congressional power. 30 7 

What emerges, then, is that concurrent authority among the 

President, the Senate, and Congress is simply a pervasive fact of constitutional 

life even (perhaps, especially) in foreign affairs and that the 

text provides manifold ways for Congress to bypass the President if it 

is so determined. Still, given the history of our constitutional practices, 

Professor Tribe is not obtuse in perceiving an awkwardness in 

the idea of independent congressional agreement-making. 303 What he 


307 Professor Tribe seems to fall into confusion in arguing at a later point that Congress's 

power to override the President's veto of a state compact or agreement -renders all the 

more remarkable a constitutional interpretation that would allow Congress to make what 

amount to treaties over presidential veto." Tribe, Taking Text, supra note 3, at 1271 n.172. 

It is hard to understand this passage as anything but a non sequitur. The fact that Congress 

can override a presidential veto of a state agreement hardly makes it more surprising that it 

could do the same for a federal agreement. It should be noted, as well, that state agreements 

with foreign states would appear to be binding under international law on the 

United States. See Wright, Control, supra note 89, § 157, at 232-33 (indicating that state 

agreements that are binding under international law involve a "national responsibility"). 

308 Perhaps, limitations on Congress's power to override a presidential veto might be 

suggested by another parallel drawn from the war powers. While the historical debates 

have mostly questioned the applicability of the President's veto power to congressional 

declarations of war, the reverse might also be defended: Congress has no power to push 

the country into war over the objections of the President. Most commentators seem to 

have rejected this view, just as they have rejected the claim that declarations are not subject 

to the veto. See, e.g., Berdahl, supra note 118, at 95-96 (acknowledging both veto and 

veto override power); Schouler, supra note 304, at 137 (same). On the other hand, the 

Framers' apparent design to make war as difficult as possible to enter would provide some 

support for finding an exception to the veto override provisions. Perhaps, by analogy, a 

similar argument could be made in favor of an override exception in the agreementmaking 

context. I will not, however, pursue this point further. Other objections to Professor 

Tribe's view make consideration of this point unnecessary. 

Alternatively, assuming Congress may issue a declaration of war even over the 

President's veto, it is less clear from a textual perspective that it can also force him as 

Commander-in-Chief to bring the army into battle. President Cleveland threatened to refuse 

to mobilize the army if Congress forced an unwanted declaration of war against Spain 

upon him. See 2 Robert McElroy, Grover Cleveland: The Man and the Statesman 249-50 

(1923). Without an army in the field, presumably, there is no fighting war, Congress's 

declaration notwithstanding. Here, again, by analogy, perhaps Congress can override a 

presidential veto of a resolution approving an international agreement, but the President 

nevertheless is under no obligation to ratify the agreement and thereby to make it binding 

on the United States internationally. In any case, as to both of the %ar powers arguments, 

my own view favoring congressional primacy in foreign affairs-most dramatically evident 

in the case of the war powers-would support Congress's ultimate authority to override



[Vol. 73:1791 

fails to notice, however, is how far this concern inevitably draws him 

away from the text. As we shall see, the pedigree for the President's 

monopoly over the conduct of diplomatic affairs is rooted principally 

in history and practice and, at best, can be derived only loosely from 

the text. Yet, history and practice are forbidden domains for a strict 

original meaning textualist. If Professor Tribe's textual assault on the 

congressional-executive agreement requires resort to such extratextual 

materials, his argument testifies more to the weakness of his own 

interpretive methodology than to the proper construction of the 

Treaty Clause. 

Professor Tribe actually tells us little about his grounds for affirming 

the President's monopoly over negotiations, but his view is 

hardly unfamiliar. It is rooted in longstanding conceptions of the 

President's authority over the conduct of our foreign affairs. From the 

beginning, the President has been the "sole organ of the nation in its 

external relations, and its sole representative with foreign nations. 

30° 

9 

He thus has exclusive authority to communicate the official policy of 

the United States to foreign states and to receive official notice of 

their policy when communicated to the United States. 310 The sole organ 

power, however, does not appear in terms in the text-a point 

Professor Tribe just glides past-and this has forced advocates of presidential 

authority to affirm expansive constructions of some rather unassuming 

provisions in Article 11.311 The most widely accepted view 

looks to the President's "control of the foreign relations 'apparatus,"' 

which in turn derives principally from his express powers to appoint 

and to receive ambassadors and other public ministers. 3 12 Whichever 

attenuated argument one ultimately prefers, however, the President's 

the President's objections and the constitutional obligation of the President to fight any 

war Congress declared even over his veto. I do not claim that this affirmation of congressional 

supremacy, however, can be justified on the basis of a plain meaning construction of 

the text. In any case, I do pursue a related argument concerning the scope of the 

President's negotiating powers in the text below. 

309 10 Annals of Cong. 613 (1800) (remarks of Rep. John Marshall). Marshall, then a 

representative in the House, made these famous remarks during the debate over President 

Adams's extradition of Jonathan Robbins to Great Britain. See generally Ruth 

Wedgwood, The Revolutionary Martyrdom of Jonathan Robbins, 100 Yale L.J. 229 (1990) 

(discussing Robbins affair and its effect on debate over executive power in foreign affairs). 

310 See Corwin, The President, supra note 119, at 213-14 (discussing history of sole organ 

power); Henkin, supra note 9, at 41-42 (discussing sources of power). 

311 Even Professor Tribe acknowledges that Article II has only a "nebulous connection 

to the subject of negotiation." Tribe, Taking Text, supra note 3, at 1257. 

312 Henkin, supra note 9, at 41-42; see also Corwin, The President, supra note 119, at 214 

& 476 n.46 (quoting justification on similar lines by President Grant). The sole organ 

power may also find additional support in the President's powers to make treaties, as 

Commander-in-Chief, and to faithfully execute the law. Perhaps, it is an implication of all 

of the powers granted in Article II.




complete monopoly over foreign communications finds its real justification 

in its acceptance from the outset and in presidents' almost invariable 

insistence upon it.313 In one famous instance, President Grant 

even returned a congressional resolution expressing thanks to foreign 

governments which had sent congratulatory messages on the first 

Centennial of the nation and used his veto message as an occasion for 

a detailed lecture on the constitutional proprieties 3 14 Only the most 

wayward Congress, then, would contemplate bypassing the President 

and appointing its own agent to negotiate or ratify an international 

agreement. 

But if the sole organ power is firmly rooted in practice and precedent, 

its scope and extent remain more open to question. Would it, 

for example, rule out a congressional resolution directing the 

President to ratify an agreement 315 Given the provenance of the 

President's powers, it is probably impossible to answer this question 

except by further reference to practice, but until Congress attempts 

such a maneuver, we will have little to go on. 316 The point, however, 

is not to attempt to undermine the expansive claims that have often 

313 For an early confirmation of this view, see Jefferson's letter to the controversial 

French Minister Genet: 

[B]eing the only channel of communication between this country and foreign 

nations, it is from [the President] alone that foreign nations or their agents are 

to learn what is or has been the will of the nation; and whatever he communicates 

as such, they have a right, and are bound to consider as the expression of 

the nation, and no foreign agent can be allowed to question it. 

Letter from Thomas Jefferson to Edmond Charles Genet, Minister of France (Nov. 22, 

1793), in 6 The Writings of Thomas Jefferson 451, 451 (Paul Leicester Ford ed., New York, 

G.P. Putnam's Sons 1895). In 1799, Congress passed the Logan Act, ch. 1, 1 Stat. 613 

(1799) (codified as amended at 18 U.S.C. § 953 (1994)), which makes it a crime for any 

person to correspond with a foreign nation with an intent to influence its conduct in relation 

to a dispute involving the United States. See generally Corin, The President, supra 

note 119, at 213-14 & 476 nn.43-44 (discussing history of Logan Act and its use). 

314 For an account of this and other equally extreme incidents, see Corwin, The 

President, supra note 119, at 214 & 476 nn.45-46. 

315 In a reversal of the actual (and tragic) trend, imagine a Congress hell-bent on ratifying 

human rights treaties and a resistant President who rejects the handiwork of earlier 

occupants of his office. For commentary on actual congressional attitudes since the 

Bricker Amendment controversy, see Louis Henkin, U.S. Ratification of Human Rights 

Conventions: The Ghost of Senator Bricker, 89 Am. J. Int'l. L 341 (1995). 

316 Nevertheless, most foreign relations scholars have thought that the sole organ power 

extends even to the formulation of the policy the President is to communicate. The 

President, many have claimed, not only communicates but determines what, if anything, is 

to be communicated. See, e.g., Henkin, supra note 9, at 42, 119, 249 (-Attempts by Congress 

to instruct U.S. representatives are highly questionable as a matter of constitutional 

separation of powers, and are usually only hortatory or are likely to be treated as such by 

the President."). Under this view, Congress would have no power to oblige the President 

to communicate an instrument of ratification to foreign states. As a consequence, even if 

Congress had the temerity, or the moxie, to override a veto of a resolution directing the 

President to ratify a particular agreement, the President would almost certainly disregard



[Vol. 73:1791 

been made on behalf of the sole organ power. 3 17 It is to underscore 

just how far removed the text itself is from these doctrinal debates and 

how it becomes even more removed as the connection to negotiations 

becomes more remote. In the longstanding constitutional debates, 

textual considerations have played only a minor role, and those arguing 

for expansive presidential powers have relied heavily on our whole 

history-on what the office of the President has become, not on what 

can be gleaned from the text or even from the early conflicts and 

crises. 318 

But if Professor Tribe's first claim is problematic, his second argument 

stands on even weaker ground. He believes that by embracing 

its pretensions. On the other hand, there is some practice arguably suggesting the contrary 

view. See supra note 291. 

The President's exclusive authority to communicate with foreign states may also provide 

an argument against a congressional power to place the nation in a state of war over 

presidential objection. Perhaps a declaration of war does not become effective as an international 

act until it is communicated to the opponent, and this power belongs solely to the 

President. See supra notes 304-05 and accompanying text (discussing veto power and declarations 

of war). 

317 My own position even in this area favors shared powers in foreign affairs with congressional 

preeminence in the event of conflict, a view with strong roots, as I have sought 

to demonstrate, in the text. In this respect, however, my approach has not always been 

fully accepted. Certainly, it was not what Justice Sutherland had in mind in his expansive 

dicta in the Curtiss-Wright case. See infra note 338 and accompanying text. 

318 See, e.g., Restatement, supra note 9, § 339 reporters' note 1 (resting President's unilateral 

power to terminate treaties on what sole organ power has become "as it has developed 

over almost two centuries"); Henkin, supra note 9, at 41-45 (describing historical 

development of sole organ power). 

Were Professor Tribe to rely on the historical materials, there would be no need for 

his exhaustive mining of the text to resolve the ambiguities inherent in the Treaty Clause. 

The point of the painstaking historical exegesis in which Professor Ackerman and I engaged 

was precisely to demonstrate that the practice during the first 150 years supports 

only the exclusive reading. In this respect, Professor Tribe's argument against independent 

congressional agreement-making faces precisely the same dilemma as his argument against 

the congressional-executive agreement. Neither of these practices are, as Professor Tribe 

would have it, inconsistent with plausible readings of the text. The principal argument 

against both is that they are inconsistent with historical understandings and practice. But 

the textual case for the congressional-executive agreement cannot be undermined by its 

purported inconsistency with the historically-based injunction against independent congressional 

agreement-making. This is especially so because the two histories are intimately 

intertwined. At least one obvious reason for the absence of any practice during the first 

150 years supporting an independent congressional agreement-making power is that the 

Treaty Clause itself was understood to be exclusive. If the Treaty Clause makes senatorial 

consent the exclusive procedure for concluding international agreements, then, a fortiori, 

Congress cannot make agreements on its own authority, either by appointing its own agent 

to conduct negotiations or by directing the President to negotiate or ratify an agreement to 

which he objects. Thus, the same practice that supports the exclusivity of the Treaty Clause 

likewise supports the case against independent congressional agreement-making. It makes 

no sense to eschew reliance on this history when arguing the main point in contention but 

then to resort to the very same history to establish a related point which can then be used 

as an argument in favor of the exclusive view.




the congressional-executive agreement, one is somehow logically estopped 

from rejecting any limitations on Congress's independent 

agreement-making powers. In his view, the arguments against Congress's 

independent powers become unavailable once one accepts a 

congressional power to approve presidentially negotiated agreements. 

Indeed, he seems to think that affirming Congress's power to approve 

agreements under Article I somehow precludes one from finding limits 

anywhere in Article H on any of Congress's Article I powers. Why 

this should be so is quite unclear-except that he (wrongly) attributes 

this view to Professor Ackerman and me, thinking he has, as it were, 

hoisted us upon our own petard. 3 19 Be that as it may, defeating a bad 

argument does not make an affirmative case, and so Professor Tribe 

would seem obliged to offer something else in support of his position. 

The best I can discern, however, is that he simply thinks it implausible 

to find independent authority in Article I for approving agreements 

despite the Treaty Clause without at the same time ruling out any 

other Article II limits on congressional powers. 

With all due respect, the reason why Professor Tribe perceives 

this as implausible escapes me entirely. First, once we recognize that 

the sole organ power itself arises principally from historical practice, 

not text, then Professor Tribe's objection becomes obtuse. Congress 

may approve agreements but for historical reasons it is enjoined from 

319 For Professor Tribe's argument, see Tribe, Taking Text, supra note 3, at 1251, 1253- 

54, 1256-58. He repeatedly invokes a brief line in our article in which we stated that Articles 

I and f1 are "'great and independent grants of power,'" id. at 1253 (quoting Ackerman 

& Golove, supra note 5, at 920 (emphasis added by Professor Tribe)), and he insists that 

this means, in our view, that Congress's and the President's powers are to be construed 

wholly independently of each other. Of course, we meant nothing of the kind, and he only 

reaches this conclusion by lifting the quotation out of context. What we said was that 

Articles I and II are "great and independent grants of power, ead of whidh suffices to 

justify the creation of international obligations." Ackerman & Golove, supra note 5, at 920 

(emphasis added). No claim of global independence was stated or implied, and any such 

position would, in my view, be wholly unpersuasive. Our only claim-and the claim I 

defend here-is that there is a plausible textual case for the view that Congress has an 

Article I power to approve international agreements that is independent of the Article II 

power of the President and the Senate to make treaties. In other areas, the President's 

Article II powers may well be best read to limit Congress's Article I authority, and vice 

versa. Such questions can only be resolved on a case-by-case basis. That Professor Tribe 

nevertheless seriously believes that his construction of our position is essential to the argument 

for the congressional-executive agreement is strongly suggested by his later argument 

that construing the Treaty Clause as nonexclusive requires one to construe the Appointments 

Clause as nonexclusive as well, hence allowing Congress to vest the appointment of 

a new Trade and Commerce Secretary in the Supreme Court. See Tribe, Taking Text, supra 

note 3, at 1273-75. Why this should be the case is mysterious. Presumably, moreover, he 

thinks acceptance of the congressional-executive agreement necessarily wipes out the 

"sole" in the sole organ power-if Congress wishes to conduct our foreign affairs, it may 

do so without intruding on the President's exclusive authority. Indeed, unless the Treaty 

Clause is exclusive, there is no exclusively executive authority.



[Vol. 73:1791 

acting independently in the realm of foreign negotiations. Where is 

the inconsistency Moreover, even were we to accept the claim, arguendo, 

that the text somehow grounds the sole organ power, Professor 

Tribe concedes that it derives principally from the President's control 

of the foreign relations apparatus through the Appointments 

Clause. 320 He is thus reduced to claiming that there is something untenable 

about some Article II powers being held (wholly or partially) 

concurrently with Congress and others entirely by the President alone. 

But why would this be untenable It certainly describes our actual 

constitutional practices. 321 And it seems consistent with common 

sense: Depending upon the nature of the power in issue, the possibly 

overlapping grants to Congress, and the context, some provisions will 

be exclusive, some nonexclusive, and others a complex mix. 3 22 There 

is, then, nothing bizarre or incomprehensible about some provisions in 

Article II being deemed exclusive while others are deemed nonexclusive. 

That is commonly the case because of the complex ways different 

constitutional provisions interact with one another. 

A moment's reflection on the two kinds of exclusivity in issue 

here, moreover, reveals that they pose radically different questions. 

Whether the Treaty Clause excludes a congressional power to approve 

international agreements depends largely, as Professor Tribe acknowledges, 

on federalism considerations-to what degree does the twothirds 

rule reflect a design to protect minority state interests. 3 2 3 In 

contrast, whether the President's control over the appointment and 

reception of diplomatic agents-the source of the sole organ powerimplies 

an exclusive presidential power to negotiate and ratify international 

agreements does not implicate federalism except in the most 

remote sense. The sole organ power raises classical problems of the 

separation of powers in foreign affairs. We should neither be surprised 

nor anxious, then, about the prospect of the two questions being 

decided in entirely dissimilar ways. Indeed, one could only miss 

this point by failing to apply a key teaching of Professor Tribe's own 

constitutional methodology-when we interpret the Constitution, we 

must take both text and structure seriously. Hence, when we consider 

whether the Treaty Clause is exclusive of a congressional power to 

approve agreements, we ask, inter alia, whether federalism so requires; 

when we consider whether the Article II sole organ power ex- 

320 See Tribe, Taking Text, supra note 3, at 1256 (citing Henkin, supra note 9, at 45-46). 

321 See supra notes 117-23, 131-34, 194-209, and accompanying text. 

322 See supra note 119. Especially revealing are the decisions in which the Court has 

held that Congress's implied powers over immigration are (partially) concurrent with the 

President's implied powers over the same subject. See supra note 120. 

323 See supra notes 68-75, 186, 273, and accompanying text; infra note 378.



1895 

clusively grants the President the power to negotiate agreements, we 

ask, inter alia, about the proper roles of the President and Congress in 

conducting our foreign affairs. 324 

324 Professor Tibe hypothesizes and then rejects a possible response to his veto override 

argument. See Tribe, Taking Text, supra note 3, at 1253 n.10S. Someone, he imagines, 

might point out that the President could always unilaterally terminate a congressionalexecutive 

agreement approved by Congress over his veto. It would thus be in his power to 

put an end to the agreement at any time, including at the moment it becomes effective. 

According to Professor Tribe, however, this argument is wrong because, although the 

President may unilaterally terminate a treaty, he may terminate a congressional-executive 

agreement only with Congress's consent. This conclusion follows, he thinks, from the general 

principle that the President may not repeal an act of Congress. See id. While I fully 

agree that the hypothetical rejoinder he postulates is wrong, there is no reason to rest its 

refutation on any limitations on the President's unilateral termination powers, a subject 

which is far more complicated than Professor Tribe acknowledges. See supra note 180. 

Perhaps his concern about the President repealing an act of Congress simply reveals a 

misconception of what terminating an international agreement involves. Mhen the 

President terminates a treaty, he is taking an act on the international level-abrogating the 

treaty for purposes of international law. The treaty ceases to have effect as domestic law 

not because the President has the power to repeal domestic law, but because the domestic 

effect of the treaty depends upon its status under international law. Similarly, when the 

President terminates a congressional-executive agreement, he is acting solely on the international 

level-upon termination, the agreement will no longer bind the United States 

under international law. But the President has no power to "repeal" the domestic effect of 

the congressional legislation implementing the agreement. Whether that legislation will 

still be binding as domestic law will depend on whether Congress explicitly or implicitly 

provided that it would survive the abrogation of the international obligation the legislation 

was designed to implement. Ordinarily, we presume that Congress intends to make such 

legislation dependent upon the reciprocally binding international obligation. Of course, 

the same applies to non-self-executing treaties that Congress implements through legislation. 

When the President terminates such a treaty, we ordinarily presume that Congress 

did not intend for the implementing legislation to continue in force notwithstanding the 

termination of the international obligation. 

In any case, there are two conclusive objections to Professor Tribe's hypothetical 

response even assuming, contrary to his claim, that the President may unilaterally terminate 

a congressional-executive agreement. F'rst, Congress will already have bound the 

nation's good faith. The fact that the President could willy-nilly place us in violation of our 

international obligations hardly cures the supposed constitutional defect. Although some 

treaties may be terminable at will, most have termination clauses restricting the rights of 

the parties to denounce or withdraw from the agreement. See Restatement, supra note 9, 

§ 332 (describing international law of treaty termination); id. § 332 cmt. a (elaborating on 

same). Second, as explained above, Congress could limit the President's termination 

power simply by writing restrictions into the resolution approving the agreement, just as 

the Senate may when giving its consent to a treaty. See id. § 339 cmt. a (affirming Senate's 

power to restrict unilateral treaty terminations); id. § 339 reporters' note 3 (elaborating on 

same); Tribe, Taking Text, supra note 3, at 1253 n.108 (suggesting same). Consequently, 

even if the President could otherwise unilaterally terminate the agreement, Congress can 

remove that power. Of course, even though Professor Tribe's hypothetical rejoinder is 

unpersuasive, we have already seen that his veto override argument is meritless on other 

grounds.



[Vol. 73:1791 

2. Fitting Square Pegs into Round Holes: 

The President's Power to Conclude Unilateral Executive 

Agreements and the Compact Clause 

One might have expected that Professor Tribe would be particularly 

concerned to protect the Treaty Clause from incursions no matter 

what their source. Yet, although he denounces the congressionalexecutive 

agreement, he is surprisingly restrained when it comes to 

another apparent assault on the Treaty Clause-the President's power 

to conclude agreements solely on his own constitutional authority. 

Rather than boldly pursuing a slash and burn expedition against all 

alternatives to the treaty form, Professor Tribe accepts the validity of 

the unilateral executive agreement heedless of the serious problems 

this creates for his own position.32- 

There are two fundamental respects in which acceptance of the 

President's unilateral powers tends to affirm the congressionalexecutive 

agreement. First, and most obviously, the unilateral agreement 

directly defeats the claim that the Treaty Clause is exclusive, 

leaving Professor Tribe's expressio unius argument up in the air. Recall 

that it was precisely the Treaty Clause's failure to include language 

providing for alternative methods of approving agreements that 

allegedly justified an inference of exclusivity. If the force of the expressio 

unius inference is not sufficient to rule out presidential agreements, 

however, then it ought not preclude congressionally approved 

agreements either, especially because the strength of the inference is 

so much greater in the former case. To the extent that presidents may 

conclude agreements on their own steam, the Treaty Clause is correspondingly 

rendered a virtual nullity. 326 In contrast, as we have seen, 

the same cannot be said for congressional-executive agreements, 

which can be harmonized with senatorial advice and consent by understanding 

the latter as an extraordinary option accorded the 

President. Furthermore, acceptance of presidential agreements substantially 

undermines the federalism argument for the exclusive read- 

325 For the distinctly negative bearing of the Appointments Clause on the textual case 

for the unilateral executive agreement, especially under Professor Tribe's formalistic interpretive 

approach, see supra note 161. This difficulty seems not to trouble him, even while 

he argues that the Appointments Clause rules out the congressional-executive agreement. 

See supra Part III.A.2. 

326 In theory, the Framers might still have had a reason for the Treaty Clause-treaties 

supersede prior conflicting law, whereas unilateral executive agreements probably do not. 

See infra note 357 and accompanying text. However, this difference is hardly adequate to 

explain the special precautions the Framers wrote into the text. Had their concern been 

solely with the domestic legal status of agreements, surely they would have left implementation 

to the whole Congress, as they in fact did for the power to pass legislation implementing 

treaties and superseding them for domestic law purposes.




ing. That argument rests on the Senate's participation in approving 

treaties. Of course, the Senate also participates in the approval of 

congressional-executive agreements. In contrast, when the President 

unilaterally concludes an agreement, the states are left entirely in the 

lurch. 327 

The second argument requires a closer look at the underlying textual 

justification for the unilateral agreement. The most obvious and 

compelling point of departure is the implied powers doctrine. In Article 

I, that doctrine is written directly into the text in the Necessary and 

Proper Clause. But as we have seen, both Madison and Marshall 

viewed it as a necessary correlate of any constitutional grant of power 

and affirmed that it would apply independently of the text. 3 28 

Although they were considering Congress's powers, the principle applies 

with equal force to the substantive powers of the President. 

Among others things, he is the sole organ and chief diplomatic representative 

of the nation and its Commander-in-Chief. Incident to these 

powers, then, he may conclude unilateral agreements as necessary to 

carry out his constitutional responsibilities. The extent of his 

agreement-making powers, in turn, depends upon the scope of his 

substantive powers. This reading not only has the merit of providing a 

strong basis in the text for the practice, it is how the President's powers 

have been widely understood from the beginning. 3 9 But once we 

327 To be sure, it is possible that, notwithstanding these arguments, the Framers might 

have accepted the President's unilateral powers as a necessary incident of his supervision 

of the conduct of foreign affairs but have rejected the same for Congress. Perhaps, the 

total exclusion of the states from unilateral presidential agreements did not concern them 

overly because they assumed that unilateral agreements would be quite modest in scope. 

Perhaps, in contrast, they thought that a parallel power in Congress to approve agreements 

incident to its enumerated powers would be more worrisome because of its potentially 

greater scope, despite the participation of the Senate by majority vote in the approval 

process. The problem for Professor Tribe, however, is that the Framers failed to write this 

distinction into the text and, for present purposes, that is the only point that matters given 

his insistence that the rarefied textual case--divorced from history-is all that is in issue. 

For further discussion of this argument, see infra note 380. 


329 See Corwin, President's Control, supra note 119, at 116-25 (treating various examples 

of unilateral executive agreements as exercises of President's diplomatic and 

commander-in-chief powers); Samuel B. Crandall, Treaties: Their Making and Enforcement 

§§ 56-61 (2d ed. 1916) (same); 2 Charles Cheney Hyde, International Law. Chiefly as 

Interpreted and Applied by the United States §§ 507-09 (1922) (providing examples of 

various classes of unilateral executive agreements); Treaties, 2 Wharton Digest § 131, at 12 

("Matters exclusively of Executive discretion ... may be settled by protocols which... 

need not be submitted to the Senate."); Wright, Control, supra note 89, §§ 161-172 (relating 

various classes of executive agreements to particular foreign affairs powers of President); 

Moore, supra note 20, at 389-92 (treating President's authority to conclude 

agreements as incident to his diplomatic powers). For further discussion and the citation of 

numerous additional early authorities, see Ackerman & Golove, supra note 5, at 815-20. 

Thus, in the early period, the President could settle private claims against foreign powers



[Vol. 73:1791 

and enter modi vivendi as incidents of his diplomatic powers, and he could conclude armistices, 

agreements for exchanges of prisoners of war, and arrangements allowing foreign 

troops to cross the border, all as incidents of his powers as Commander-in-Chief. See id. 

His powers, however, were narrowly construed. His diplomatic powers, for example, did 

not allow him to enter political agreements that were binding on the country; they were 

thought to bind only the President that made them. See id. at 819-20. The Supreme Court 

reflected these early understandings in Tcker v. Alexandroff, 183 U.S. 424 (1902), in 

which, after reviewing the past precedents in which the President had entered agreements 

permitting foreign troops to enter the United States, it noted that while no act of Congress 

authorized this practice, "the power to give such permission... was probably assumed to 

exist from the authority of the President as commander-in-chief." Id. at 435. 

With the emergence of the United States on the world stage in the wake of the 

Spanish-American War, the President's unilateral foreign affairs powers experienced a corresponding 

expansion which was reflected in the increased use of the unilateral agreement. 

This development provoked considerable constitutional controversy. See Ackerman & 

Golove, supra note 5, at 817-18 & 818 n.64. In an effort to cabin this practice, which could 

no longer be easily contained by the implied powers doctrine, some commentators looked 

to the Compact Clause's distinction between treaties and agreements or compacts as a 

suggestive analogy, revealing that the Framers were at least aware that international agreements 

assumed different forms. This suggested to some that the President's unilateral 

powers might be limited to lesser forms of international undertakings. See, e.g., Simeon E. 

Baldwin, The Exchange of Notes in 1908 between Japan and the United States, 3 Zeitschrift 

fur Volkerrecht und Bundesstaatsrecht 456, 457-61, 464-65 (1909) (arguing that 

President may unilaterally conclude only agreements that do not rise to level of treaties); 

James F. Barnett, International Agreements Without the Advice and Consent of the Senate, 

15 Yale L.J. 18, 63, 82 (1905) (same); see also 2 Hyde, supra, § 505, at 27 (opening 

discussion of unilateral executive agreements with observations that Framers were familiar 

"with the habit of statesmen to contract international engagements of lesser importance 

than treaties"). Even for these scholars, however, the point was to find an additional limitation 

on the President's authority, not to argue against the implied powers doctrine, which 

was accepted uniformly. See, e.g., Barnett, supra, at 70-73 (invoking Commander-in-Chief 

Clause to justify President's authority to conclude armistices, peace protocols, and joint 

military occupation agreements). The lack of any persuasive standards by which to define 

the difference between treaties and agreements, moreover, bedeviled the argument from 

the outset. See, e.g., Baldwin, supra, at 457-59 (resting on Vattel for distinction); see also 

infra note 361 (discussing problematic character of Vattel's distinction for purposes of limiting 

President's unilateral authority). In any case, the Compact Clause played only a minor 

role in the debates, see, e.g., 2 Butler, supra note 47, § 463, at 367 n.2 (failing to 

mention Compact Clause in discussion of scope of President's unilateral power to make 

agreements); Corwin, President's Control, supra note 119, at 116-25 & 121 n.48 (mentioning 

Compact Clause only in passing in brief footnote and making nothing of distinction); 

Crandall, supra, §§ 56-61 (failing to mention Compact Clause in discussing limits on 

President's implied powers to make agreements); Wright, Control, supra note 89, §§ 161- 

172 (same); and it failed to appear at all when the Supreme Court considered the constitutionality 

of the practice in a series of later cases. In each, the Court, not mentioning the 

Compact Clause, upheld the agreements as properly incident to the President's foreign 

affairs powers. See United States v. Belmont, 301 U.S. 324, 330-31 (1937) (upholding 

Litvinov Assignment, concluded by President Roosevelt as part of series of measures leading 

to recognition of U.S.S.R, as incident to President's sole organ and recognition powers, 

deriving latter power from his power to receive ambassadors); United States v. Pink, 315 

U.S. 203, 228-30 (1942) (same); Dames & Moore v. Regan, 453 U.S. 654, 682-83 (1981) 

(following Pink and upholding Iranian Hostages Accord, which settled claims with Iran 

and referred them for resolution to Iran-U.S. Claims Tribunal, as incident to President's 

sole organ and recognition powers). Although the Compact Clause has never dropped



1899 

accept agreement-making as an implied power of the President, we 

have taken a giant leap towards acceptance of the congressional-executive 

agreement, since nothing in the Treaty Clause provides a textual 

foothold for distinguishing between the implied powers of the President 

and the Congress in this regard. 

Professor Tribe thus faces a perplexing predicament-having 

committed himself to the unilateral presidential agreement, how can 

he avoid breathing new life into the congressional-executive agreement, 

its close constitutional relation This is surely a difficult challenge, 

but he offers a three-part argument. Although designed to 

underwrite his plain meaning construction of the text, however, his 

claims begin with the controversial, proceed to the doubtful, and end 

in the highly dubious. 

Professor Tribe begins with the apparent discovery of a genuine 

hole in constitutional space. He notes that although the Treaty Clause 

may appear on its face to be a comprehensive grant of the agreementmaking 

power to the President and the Senate, in fact the Framers 

used the word "treaties" as a term of art 33 0 The power to make "treaties" 

does not include the power to make all international agreements, 

only important agreements that substantially impinge on state or national 

sovereignty. 3 31 To demonstrate why, he points to Article I, Section 

10. The Framers prohibited states from entering into treaties, 

alliances, or confederations, but allowed them, with Congress's consent, 

to enter into agreements or compacts (agreements, in his view, 

that less substantially impinge on state or national sovereignty) 332 

Since he believes that the term "treaty" has the same meaning 

throughout the text, it is clear that the power to make treaties does 

not comprehend the power to make agreements or compacts. 3 33 We 

thus face a potential gap in constitutional space-the Framers were 

silent on how less important agreements and compacts would be made 

entirely from view, this approach is now widely accepted. The Restatement, for example, 

provides: "[T]he President, on his own authority, may make an international agreement 

dealing with any matter that falls within his independent powers under the Constitution." 

Restatement, supra note 9, § 303(4); see also Congressional Research Serv., Library of 

Congress, 103d Cong., Treaties and Other International Agreements: The Role of the 

United States Senate, 60-65 (Comm. Print 1993) [hereinafter Congressional Research 

Serv., Treaties] (authoritative study prepared for Senate Foreign Relations Committee) 

(basing President's authority on various Article II powers); Restatement, supra note 9, 

§ 303 cmt. g & reporters' note 11 (same). For further discussion, see supra notes 64-64 and 

accompanying text; infra note 376. 

330 See Tribe, Taking Text, supra note 3, at 1265. 

331 See id. 

332 See id. at 1265-66. 

333 See ic at 1266.



[Vol. 73:1791 

on behalf of the United States (although oddly they provided explicitly 

for how they might be concluded by states). 

Fortunately, Professor Tribe has a solution to the problem he has 

uncovered. As Curtiss-Wright explained, the foreign affairs powers of 

the national government are plenary, and so the power to make less 

important agreements must necessarily reside in some part of the federal 

government. 334 According to Professor Tribe, this power clearly 

falls to the President for two reasons. First, following Hamilton, the 

Executive Power Clause can be read as a grab-bag grant of any and all 

powers that are "executive" in character. 335 Thus, in contrast to Congress, 

the President is not limited to his enumerated powers. Second, 

pace Hamilton in The Federalist, the making of agreements is executive 

in character. It therefore belongs to the President, not to 

Congress. 336 

To see the many difficulties in this argument, we must examine 

each of its parts separately. Begin with his explicit reliance on the 

claim that agreement-making is an executive function. We confronted 

this assertion in a previous section, and I pointed out then that the 

text of the Constitution, as well as the evidence of the Framers' contemporary 

understandings, better supports a more nuanced conclusion: 

While the negotiating of treaties is an executive function, their 

approval is ordinarily a legislative function. 3 37 Professor Tribe cites no 

textual or historical authority in favor of his view. Indeed, the only 

evidence he cites is the Curtiss-Wright case. But Justice Sutherland's 

opinion, though a paean to executive leadership in foreign affairs, says 

nothing at all, even in its most notoriously expansive dicta, about the 

334 See United States v. Curtiss-Wright Export Corp., 299 U.S. 304, 315-18 (1936) 

(claiming that foreign affairs powers are plenary and are vested in federal government as 

an incident of national sovereignty). 


336 For Professor Tribe's argument on these points, see Tribe, Taking Text, supra note 3, 

at 1264-69. 





specific character of the power to approve agreements. 3 Hence, one 

essential predicate of Professor Tribe's argument is doubtful at best. 3 39 

Second, Professor Tribe self-consciously rests on the Hamiltonian 

reading of the Executive Power Clause. 40 Although he recognizes 

the controversial character of that view, he is forced to this position 

for at least two reasons. On the one hand, as I have already pointed 

out, were he to concede that presidential authority can be implied 

from enumerated grants in Article II, he would have no persuasive 

response to the advocate of congressional power who seeks to make 

the same move under Article I. But were he then to concede that 

Congress can approve some agreements under the Necessary and 

Proper Clause, he would, at a minimum, have to abandon one of his 

central claims: that a resolution approving an agreement is categorically 

not a "law" for purposes of that clause.3"4 Equally problematic, 

the implied powers doctrine cannot fully solve the dilemma he has 

created, for it is unable to fill entirely his postulated constitutional 

gap. Implied powers permit the conclusion of agreements that are incident 

to an enumerated power. But Professor Tribe needs a general 

undifferentiated power to make all agreements falling below a certain 

threshold of importance, irrespective of their relationship to an enumerated 

power. Especially if we consider the far more constricted 

conception of executive powers held by the Framers, implied powers 

may not be up to the task. Hence, Professor Tribe must unearth an 

unenumerated power to fill in the void. In this context, Hamilton's 

thesis is understandably alluring. But to say that Hamilton's position 

338 Indeed, Justice Sutherland's only reference to the power to make agreements without 

the consent of the Senate cites the Court's earlier decision in B. Altman & Co. v. 

United States, 224 U.S. 583, 600-01 (1912). See Curtiss-Wright, 299 U.S. at 318. AItman, 

however, dealt with a precursor to the modem ex ante congressional-executive agreement. 

See Ackerman & Golove, supra note 5, at 831-32 (discussing Aitman). If anything, then, 

Curtiss-Wright would appear to support the claim that approving agreements is at least 

sometimes a legislative function. 

339 This is not to suggest that concluding some types of agreements might not be thought 

of as properly executive in character. Despite the lack of any specific evidence about the 

Framers' view, one might reasonably think that this characterization would apply to agreements 

"as to the rights of an individual, the treatment of a vessel, a matter of ceremonial, 

or any of the thousand and one things that daily occupy the attention of foreign offices 

without attracting public notice." Moore, supra note 20, at 389. But no one-and certainly 

not Professor Tribe-claims that the President's unilateral authority is limited to these routine 

administrative matters. See id. at 390-93 (providing examples of more substantive 

agreements concluded by President acting alone). 

3M See supra notes 106-09, 299, and accompanying text. 

341 For further discussion, see infra note 378. Furthermore, Professor Tribe would seriously 

weaken any argument for placing limits on the kinds of agreements Congress can 

approve incident to its foreign affairs powers, since nothing in Article I, Section 8 or the 

Necessary and Proper Clause suggests that the Framers wished to allow Congress to approve 

some agreements but not others. See infra note 378.


1902 


[Vol. 73:1791 

is controversial profoundly understates the longstanding historical 

controversy, and it remains one of the great open questions of constitutional 

law. 342 To the extent that the case for the exclusivity of the 

Treaty Clause rests on an expansive reading of the Executive Power 

Clause, this merely underscores the indeterminacy of the Constitution 

on the very point in issue. All the more so, then, when the text offers 

a reasonable, and relatively uncontroversial alternative: In accordance 

with the traditional understanding, the President makes unilateral 

agreements, not pursuant to some unenumerated power, but as 

an incident of his enumerated foreign affairs powers. 3 43 

Most problematic of all, however, is Professor Tribe's central 

claim that the Constitution creates two kinds of agreements, important 

treaties and less important agreements, and that the Treaty 

Clause is limited to the former. It is easy to see why he finds this 

argument appealing. By positing the existence of two separate kinds 

of agreements and a gap in the text, he can preserve his claim that the 

Treaty Clause is exclusive within the range of its application. To be 

sure, he would say, the President may conclude unilateral executive 

agreements, but that is a separate power not comprehended by the 

Treaty Clause. When he wishes to make a treaty, the President is 

obliged to follow the singular procedure specified in the text for important 

agreements: senatorial advice and consent. Thus, there is no 

inconsistency between the presidential agreement and the expressio 

342 Hamilton initiated the controversy in his Pacificus letters which immediately 

prompted Madison's famous refutation (at Jefferson's urging) as Helvidius. See Henkin, 

supra note 9, at 39-40 & 338 n.13 (discussing historical controversy); supra notes 106-09 

and accompanying text (discussing Hamiltonian view). For Hamilton's argument, see 

Pacificus No. 1, supra note 106, at 38-39. Webster and Calhoun later opposed Hamilton's 

view. See Henkin, supra note 9, at 338 nn.13 & 17. And while Chief Justice Taft seemed to 

adopt it in Myers v. United States, 272 U.S. 52, 118, 128 (1926), Justice Black, writing for 

the Court, pointedly refused even to consider the Executive Power Clause as a possible 

source of presidential power in the Steel Seizure Case, Youngstown Sheet & Tube Co. v. 

Sawyer, 343 U.S. 579, 587-88 (1952). In his celebrated concurring opinion, moreover, Justice 

Jackson, President Roosevelt's former Attorney General, expressly denounced such a 

doctrine as dangerous, and the Court has not invoked it since. See id. at 640-41 (Jackson, 

J., concurring). Academic opinion is also skeptical. See, e.g., Henkin, supra note 9, at 39- 

40 (discussing argument that Executive Power Clause is source of foreign affairs powers); 

Edward S. Corwin, The Steel Seizure Case: A Judicial Brick Without Straw, 53 Colum. L. 

Rev. 53, 53-55 (1953) (warning that overextension of presidential power through 

Hamiltonian reading of Executive Power Clause threatens rule of law). Ironically, one the 

most powerful objections to Hamilton's position is its textual implausibility. Why would 

the Framers have enumerated trivial powers, like the power to require written opinions by 

the heads of executive departments, if the Executive Power Clause was a grant in bulk of 

all powers that are executive in character See supra note 109 (discussing objections to 

Hamiltonian view). 

343 See supra notes 328-29 and accompanying text; infra note 376.




unius and federalism arguments for the exclusivity of the Treaty 

Clause. 

No doubt this argument has a nice elegance. But something more 

than elegance is needed to sustain Professor Tribe's plain meaning argument. 

He must articulate a construction of the text that is at least 

compelling, if not singularly persuasive. Yet, the gap which he purports 

to uncover is really only a gap of his own creation, and his claim 

that the President and the Senate are limited to making important 

agreements is based on the creative importation of implicit limits on 

an otherwise unqualified grant of power. As we shall see, his view is 

also demonstrably in conflict with assumptions that have been widely 

held during our whole history: that the Treaty Clause is an unrestricted 

grant of authority to the President and the Senate to make 

any international agreement on any subject touching upon our foreign 

affairs. But if there is no gap-if the President and the Senate can 

make any and all agreements no matter how important or unimportant-then 

there is no reason to repair to the Executive Power Clause 

to find an unenumerated power to make unimportant agreements and 

no gainsaying the fact that accepting the unilateral executive agreement 

renders the Treaty Clause nonexclusive. In short, Professor 

Tribe's effort to preserve the consistency of his position falls apart. 

Professor Tribe is not the first to have suggested limits on the 

scope of the treaty power. Many have feared the potentially awesome 

power accorded to the President and the Senate in textually unrestricted 

terms and so have sought to find implicit limits. 3 " But in all 

the historical controversy, the sweeping scope of the power has only 

rarely been seriously denied, and even Professor Tribe is satisfied in 

this regard to let things rest. Instead, he takes an entirely different 

tack: Although the Senate may bind the country to the most momentous 

of commitments, there are some agreements, he believes, that in 

their triviality are simply beyond its power of advice and consent. This 

view, however, is confronted with an immediate difficulty, for it conflicts 

with the widely held assumption from the beginning that the 

344 Henkin provides a particularly useful discussion. See Henkin, supra note 9. at 185- 

98; see also H.R. Rep. No. 48-2680, at 1-6 (1885) (arguing for limits on scope of treaty 

powers); 1 Butler, supra note 47, §§ 3-7 (discussing early controversies and taking expansive 

view of treaty power); 2 John Randolph Tacker, The Constitution of the United States 

§§ 354-356 (Chicago, Callaghan & Co. 1899) (same, but taking narrow view); 1 Westel 

Woodbury Willoughby, The Constitutional Law of the United States §§ 206-219 (1910) 

(same, taking expansive, but more moderate view). There have been arguments for restrictions 

derived variously from the nature of international undertakings, the specific prohibitions 

of the Constitution, the principle of federalism, and the separation of powers doctrine 

in its many different guises. Some of these have been accepted, but most have been rejected 

in time. See Henkin, supra note 9, at 185-98 (providing critical review).



[Vol. 73:1791 

Treaty Clause is a comprehensive grant of the agreement-making 

power. Justice Story's early dictum is typical: "The power 'to make 

treaties' is by the constitution general; and of course it embraces all 

sorts of treaties, for peace or war ... and for any other purposes, 

which the policy or interests of independent sovereigns may dictate in 

their intercourse with each other. '345 This same judgment, repeated 

countless times by presidents, secretaries of state, members of congress, 

and scholars, finds authoritative expression in the Restatement: 

"[T]he President, with the advice and consent of the Senate, may 

make any international agreement of the United States in the form of 

a treaty. '3 46 Even the Supreme Court has abandoned its traditional 

345 3 Story, supra note 202, § 1502, at 355. 

346 Restatement, supra note 9, § 303(1); see also id. § 303 cmt. b (elaborating on same). 

Hamilton reached the same conclusion, noting: 

It was impossible for words more comprehensive to be used than those 

which grant the power to make treaties. They are such as would naturally be 

employed to confer a plenipotentiary authority.... With regard to the objects 

of the Treaty, there being no specification, there is of course a charte blanche. 

The general proposition must therefore be that whatever is a proper subject of 

compact between Nation & Nation may be embraced by a Treaty .... 

The Defence No. 36, supra note 47, at 6; see also Congressional Research Serv., Teaties, 

supra note 329, at xxxviii, 41-42 (noting that "treaty power is recognized by the courts as 

extending to any matter that is properly the subject of international negotiations"); 29 

Annals of Cong. 531-32 (1816) (remarks of Sen. Calhoun) (noting that "[w]hatever, then, 

concerns our foreign relations; whatever requires the consent of another nation, belongs to 

the treaty power"); 11 U.S. Dep't of State, Foreign Affairs Manual § 721.2(a) (rev. ed. Feb. 

25, 1985) ("The President, with the advice and consent of two-thirds of the Senators present, 

may enter into an international agreement on any subject genuinely of concern in 

foreign relations ...."), reprinted in Congressional Research Serv., Treaties, supra note 

329, at 301, 303; 1 Butler, supra note 47, § 3, at 5 (stating that treaty power "extends to 

every subject which can be the basis of negotiation and contract between any of the sovereign 

powers of the world"); 1 id. §§ 261, 264, 268 (discussing views of leading early commentators 

on Constitution, including William Rawle, George Ticknor Curtis, and John 

Norton Pomeroy, which were to same effect); Treaties, 5 Hackworth Digest § 462, at 5-11 

(collecting various statements, including by Secretary of State Elihu Root, Charles Evans 

Hughes (later Secretary of State and Chief Justice), Quincy Wright, and Senator Kellogg 

(later Secretary of State)); Henkin, supra note 9, at 185-99 (affirming unrestricted character 

of treaty power); Treaties, 5 Moore Digest § 735, at 164 (reprinting note by Secretary of 

State Calhoun asserting that "treaty-making power has, indeed, been regarded to be so 

comprehensive as to embrace, with few exceptions, all questions that can possibly arise 

between us and other nations, and which can only be adjusted by their mutual consent"); 

Wright, Control, supra note 89, §173, at 24748 (noting that treaty power "extends to 'any 

matter which is properly the subject of negotiations with a foreign country"' (quoting 

Geofroy v. Riggs, 133 U.S. 258, 266 (1890))); Charles Cheney Hyde, Constitutional Procedures 

for International Agreement by the United States, 31 Am. Soc. Int'l L. 45,53 (1937) 

(affirming same); Moore, supra note 20, at 388 (same). There is some isolated language in 

a few of the early twentieth century commentators that could be read to suggest otherwise, 

but in context it is apparent that they were seeking to explain the basis for and limits on the 

President's unilateral powers, not to suggest restrictions on the scope of the treaty power. 

See Barnett, supra note 329, at 18, 82; Moore, supra note 20, at 388. John Bassett Moore, 

for example, declared: "[lilt can easily be demonstrated that the word 'treaties,' as used in




reticence about foreign affairs and repeatedly joined the chorus.3 47 

the constitutional law of the United States, does not embrace any and every kind of international 

agreement." Id. But this statement is immediately followed by his explanation 

that the term "treaty" has a purely procedural meaning, referring to "agreements approved 

by the Senate," and his affirmation that any agreement approved by the Senate is automatically 

"in the strict sense a 'treaty,"' with the specific legal character mandated for treaties 

in the Supremacy Clause. Id.; see also infra note 359 and accompanying text (discussing 

Moore's view). Moore also specifically recognized that even when an agreement falls 

within the President's unilateral powers, the Senate is not without power over the same 

agreement. See Moore, supra note 20, at 393, 414 (recognizing that President may conclude 

claims settlements and postal conventions as executive agreements or may submit 

them for senatorial advice and consent); see also Barnett, supra note 329, at 6S-70, 76-77 

(recognizing that President may conclude claims settlements, postal conventions, and tariff 

agreements as executive agreements but that Senate may approve them as treaties as well). 

347 See, e.g., Asakura v. City of Seattle, 265 U.S. 332,341 (1924) (affirming unrestricted 

character of treaty power); Geofroy v. Riggs, 133 U.S. 258,266-67 (1890) (same); Holden v. 

Joy, 84 U.S. (17 Wall.) 211, 243 (1872) (same); Holmes v. Jennison, 39 U.S. (14 Pet.) 540, 

569 (1840) (opinion of Taney, C.J.) (same). In Geofroy, the Court, per Justice Field, 

declared: 

That the treaty power of the United States extends to all proper subjects of 

negotiation between our government and the governments of other nations, is 

clear.... [I]t is not perceived that there is any limit to the questions which can 

be adjusted touching any matter which is properly the subject of negotiation 

with a foreign country. 

Geofroy, 133 U.S. at 266-67. The Court did recognize, however, that there might be limits 

on the treaty power of an entirely different order. "The treaty power, as expressed in the 

Constitution, is in terms unlimited except by those restraints which are found in that instrument 

against the action of the government or of its departments, and those arising from the 

nature of the government itself and of that of the States." Id. at 267. 

Against these clear and weighty authorities, it is arguable, although by no means clear, 

that Justice Sutherland expressed a contrary view in a passing dictum in his Curtiss-Wright 

opinion. See United States v. Curtiss-Wright Export Corp., 299 U.S. 304, 318 (1936). 

Curtiss-Wright, of course, did not involve an international agreement of any kind, dealing 

only with the scope of the nondelegation doctrine in external, as opposed to domestic, 

affairs. Nonetheless, Justice Sutherland famously took the case as an opportunity to expound 

upon his controversial theory of the foreign affairs powers. See id. at 315-22. The 

external powers of the United States, he claimed, do not depend upon the affirmative 

grants in the Constitution but simply devolve upon the federal government as concomitants 

of the concept of sovereignty under international law. Hence, "[t]he powers to declare 

and wage war, to conclude peace, to make treaties, to maintain diplomatic relations 

with other sovereignties, if they had never been mentioned in the Constitution, would have 

vested in the federal government as necessary concomitants of nationality." Id. at 318. 

But these powers were, in fact, expressly delegated to the United States, and so, to demonstrate 

his point, Justice Sutherland cited a number of other powers that are not expressly 

mentioned in the text but still, he claimed, had been found to exist "as inherently inseparable 

from the conception of nationality" and without which "the United States is not completely 

sovereign." Id. Among these, he said, was "the power to make such international 

agreements as do not constitute treaties in the constitutional sense." Id.; see also United 

States v. Belmont, 301 U.S. 324, 330-31 (1937) (arguably suggesting same in opinion by 

Justice Sutherland rendered one year after Curtiss-Wright). 

If Justice Sutherland indeed meant to suggest that there are two kinds of international 

undertakings comprising exclusive sets, treaties and agreements, and that the Senate has 

no power to consent to agreements, only to treaties-and that an unenumerated 

agreement-making power was therefore required to enable the federal government to con-



This point, moreover, is not only expressed ad infinitum in statements 

of principle, it is also reflected in the practice of the nation from 

its earliest days. For example, although presidents have often settled 

private claims unilaterally, 348 they have also frequently submitted 

these agreements to the Senate for its consent. In 1905, the State Department's 

Solicitor concluded that either method was permissible: 

"Such an agreement the president no doubt may in any case submit to 

the Senate, if he sees fit to do so; and we find, especially in former 

times, that this course was often taken. ' 349 This same view has also 

been expressed on countless occasions, including by the Supreme 

Court in Dames & Moore v. Regan 350 : "Though those settlements 

have sometimes been made by treaty, there has also been a longstanding 

practice of settling such claims by executive agreement without the 

elude agreements-there are several reasons why his view is unpersuasive. First, it is evident 

that Justice Sutherland was driven to this position not for textual or historical reasons, 

but in order to find support for his remarkable claim that the foreign affairs powers are 

inherent and not dependent on enumeration. The only decision he cites for his claim about 

the agreement-making power is B. Altman & Co. v. United States, 224 U.S. 583, 600-01 

(1912). See Curtiss-Wright, 299 U.S. at 318. Altman, however, involved a tariff reciprocity 

agreement authorized by act of Congress, and although the Court did not address the issue 

squarely, it clearly grounded the agreement's validity in Congress's foreign commerce powers 

and the President's duty to faithfully execute the law, not in the law of nations or 

sovereignty. See Altman, 224 U.S. at 601 (observing that agreement dealt with "important 

commercial relations between the two countries" "authorized by the Congress" and "proclaimed 

by the President"). Nor does anything in the opinion suggest that the tariff agreement 

could not have been submitted to the Senate-an outlandish proposition in light of 

the long national experience with tariff treaties. See, e.g., U.S. Tariff Comm'n, Reciprocity 

and Commercial Treaties 21-38 (1919) (describing history of tariff treaties and agreements); 

Barnett, supra note 329, at 64-65, 70 (referring to same and concluding "there are 

certain ends, which, under our Constitution, may be attained in two different ways"). Not 

only can Justice Sutherland find no support in Altman, the one decision on which he relies, 

he also fails even to acknowledge the long line of decisions and the weighty history of 

opinions among leading political figures and commentators that are directly to the contrary. 

See supra notes 345-46 and accompanying text. No doubt for these reasons his 

dictum either has not been so interpreted or has simply been disregarded in subsequent 

authoritative sources. See supra notes 345-46 and accompanying text. 

348 See generally Ackerman & Golove, supra note 5, at 815-20 (reviewing history and 

scope of unilateral executive agreement-making, including claims settlements). Perhaps 

the first unilateral executive agreement under the Constitution was John Adams's settlement 

in 1799 of the private claims of American citizens in the Wilmington Packet affair. 

See Wallace McClure, International Executive Agreements: Democratic Procedure Under 

the Constitution of the United States 43-44 (1941) (providing history). 

349 Moore, supra note 20, at 399. Indeed, in our first century, the preferred method 

appears to have been the treaty. See Richard B. Lillich, The Gravel Amendment to the 

Trade Reform Act of 1974, 69 Am. J. Int'l L. 837, 844-45 (1975) (noting that treaties were 

predominant form for making claims settlements in nineteenth century). In 1905, Moore 

counted 20 instances in which claims settlements had been processed as treaties; Presidents 

had settled many unilaterally as well. See Moore, supra note 20, at 402-03, 408 (compiling 

examples). 

350 453 U.S. 654 (1981).




advice and consent of the Senate." 351 On Professor Tribe's view, of 

course, Dames & Moore was wrong: If the President has unilateral 

authority to conclude these agreements, then they were ipso facto beyond 

the Senate's power of advice and consent. Indeed, on his view, 

the Litvinov Assignment 352 approved in United States v. Belmont 353 

and United States v. Pink, 354 as well as the Iranian Hostages Accord 

affirmed in Dames & Moore, both unilateral agreements, were by definition 

too trivial to fall within the ambit of the Senate's treaty 

powers! 355 

One might nevertheless respond by pointing out that although 

recourse to the Senate is unnecessary for less important agreements, 

its advice and consent in no way undermines the validity of a presidential 

agreement; it is just that the agreement remains a unilateral 

executive agreement since the Senate's consent is unable to transform 

it into a treaty. 356 Hence, the lack of concern about this problem in 

the authorities I have cited. 

This response, however, is unconvincing for at least two reasons. 

First, it simply waives away the many unambiguous statements to the 

contrary that span the course of our history. Nothing in those authorities 

betrays that they somehow sub silentio accepted Professor Tribe's 

view. Furthermore, there are important differences in the legal status 

of treaties and unilateral presidential agreements that would have required 

attention had Professor Tribe's limited reading of the treaty 

power been accepted. Most salient, in contrast to treaties, it is widely 

thought that unilateral agreements do not supersede earlier inconsis- 

351 Id. at 679. For repeated affirmations of the availability of both procedures, see, e.g., 

Crandall, supra note 329, at 86; Treaties, 5 Hackworth Digest § 515, at 404; Agreements 

Not Submitted to the Senate, 5 Moore Digest § 752, at 211 (reprinting statement of Secretary 

of State Cass); Wright, Control, supra note 89, § 171; Barnett, supra note 329, at 76-77; 

John W. Foster, The Treaty-Making Power Under the Constitution, 11 Yale LJ. 69, 77 

(1901); Lillich, supra note 349, at 844-45. 

352 See supra note 329. 

353 301 U.S. 324 (1937). 

354 315 U.S. 203 (1942). 

355 Claims settlements are not the only kinds of agreements that were alternately concluded 

as executive agreements and treaties during the first century of the country. 

Among others, postal conventions, tariff agreements, and even modi vivendi all took on 

both forms. See, e.g., 2 Butler, supra note 47, § 463, at 367 n2 (describing alternative use 

of both forms for modi vivendi); John Mabry Mathews, American Foreign Relations 436 

(1928) (describing same for agreements for United States superintendence of foreign country's 

debt payments to other foreign creditors); Barnett, supra note 329, at 6S-70 (describing 

same for postal conventions and reciprocal tariff agreements); Moore, supra note 20, at 

393 (describing alternative use of both forms for postal conventions). 

356 One might so argue, but as we shall see, Professor Tribe does not. See infra note 366 

and accompanying text.



[Vol. 73:1791 

tent laws or treaties. 357 As a result, it will not be sufficient for the 

President simply to submit an agreement falling into the gray area 

between "treaty" and "agreement" to the Senate, for if the agreement 

is really not important enough to merit the "treaty" appellation, the 

Senate's approval cannot give it this increased dignity. Prior inconsistent 

laws will still apply even in the face of the Senate's efforts. 3 58 But 

these and other similar issues have not troubled presidents, senates, 

and congresses when they have considered agreements approved by 

the Senate. They have simply assumed that any agreement is "by reason 

of its approval by the Senate, in the strict sense a 'treaty,' and 

possesses, as the product of the treaty-making process, a specific legal 

character .. overriding any inconsistent provisions not only in the 

constitutions and laws of the various states, but also in prior national 

statutes." 359 

357 The Fourth Circuit so held in United States v. Guy W. Capps, Inc., 204 F.2d 655, 660 

(4th Cir. 1953), aff'd on other grounds, 348 U.S. 296 (1955). See Congressional Research 

Serv., Treaties, supra note 329, at 65-68 (affirming this view); Tribe, Constitutional Law, 

supra note 11, § 4-5, at 229 (noting that "[a]t a minimum, it seems clear that [a unilateral] 

executive agreement, unlike a treaty, cannot override a prior act of Congress"). The Restatement 

concurs, although with nuances. See Restatement, supra note 9, § 115 cmt. c & 

reporters' note 5 (affirming same but noting possible exceptions). Professor Henkin, however, 

is less certain. See Henkin, supra note 9, at 228 (suggesting that same arguments as 

to why treaties supersede earlier statutes also apply to executive agreements). 

358 Consider also the perverse result of this view: When the Senate approves an important 

agreement, it can override any prior law even without the participation of the House. 

But when it gives its advice and consent to an unimportant agreement, its powers run out, 

and the President will have to obtain the House's consent to repeal any conflicting laws. 

This obviously reverses the intuitive notion. Nor is it an answer to suggest that any agreement 

inconsistent with prior law is automatically a "treaty" under Article II. Defining the 

difference between "treaties" and "agreements" in this way finds no support in the eighteenth 

century international practice from which, in Professor Tribe's view, the Framers 

were drawing. Professor Tribe himself interprets the point of difference as resting on the 

significance and extent of an agreement's impact on the nation's sovereignty. See supra 

text accompanying notes 334-36; supra note 161; infra note 376. Others have looked to an 

entirely different distinction found in Vattel, of which the Framers might have been aware. 

See supra note 329; infra note 361 and accompanying text. Either way, the crucial point 

has nothing to do with the agreement's relationship to preexisting law. This is not to suggest 

that a different view might not be plausible if one assumes-as I do-that the source 

of the President's unilateral authority resides in the implied powers doctrine. But if one is 

committed to the claim that the President has an unenumerated power to make all of those 

international undertakings that in the eighteenth century international practice familiar to 

the Framers were deemed "agreements" rather than "treaties," then there is no basis for 

limiting the President to agreements consistent with preexisting law. This is not to say, 

however, that a unilateral executive agreement, even under this view, would necessarily 

supersede a prior inconsistent statute. Were the President to enter into a unilateral agreement 

inconsistent with a prior statute, he would, perhaps, have to obtain repealing legislation 

before his agreement could be given effect in domestic law. 

359 Moore, supra note 20, at 388. Nor can this be explained by the invariably important 

nature of the agreements presidents have referred to the Senate Foreign Relations Committee. 

Notoriously, presidents have sometimes opted for the arguably insulting practice




Even if we overlook this history and the views of our forebears, 

Professor Tribe's position is equally problematic as a matter of pure 

textual exegesis. The plausibility of his position rests entirely on his 

assertion that the term "treaty" must have the same meaning in Articles 

I and II. It is far from clear, however, that this premise supports 

his interpretive conclusion. In any case, while this same meaning rule 

may provide a reasonable starting point for interpretation, it loses all 

force when applied woodenly without reference to contextm and 

context gives us powerful reasons for giving the term "treaty" a different 

construction in Articles I and II. 

Recall that Professor Tribe's contention is that the term "treaty" 

in Article I, Section 10 cannot include all international agreements 

because the Framers prohibited the states from making treaties but 

allowed them to make agreements or compacts. Thus, he concludes, 

treaties and agreements describe two exclusive sets. But it is at least 

equally plausible to read the permissive reference to agreements and 

of sending extremely trivial "treaties" to the Senate, while purporting to conclude rather 

hefty "agreements" on their own authority. See, e.g., S. Rep. No. 91-129, at 28 (1969) 

(noting that in some instances "we have come close to reversing the traditional distinction 

between the treaty as the instrument of a major commitment and the executive agreement 

as the instrument of a minor one"). Indeed, Senator Robert Taft complained over 50 years 

ago: "As a matter of fact, no treaties of any importance have been submitted to the Senate 

since I have been a Member of the body." 88 Cong. Rec. 9276 (1942) (remarks of Sen. 

Taft); see also 115 Cong. Rec. 16750 (1969) (remarks of Sen. Church) (echoing Sen. Taft's 

comment). 

There are a number of other legal differences between treaties and unilateral agreements 

that would also have been relevant. For example, although it is generally assumed 

that the President can modify a treaty only by obtaining the consent of the Senate, the 

same would not apply to a unilateral agreement. See Congressional Research Serv., Treaties, 

supra note 329, at 140-46 (concluding that modifications of unilateral executive agreements 

are solely matter of presidential discretion). Thus, were the response suggested in 

the text correct, the President would be free to modify unilaterally agreements approved 

by the Senate if he could demonstrate that they did not in fact rise to the level of a genuine 

treaty. Likewise, the Senate may make reservations to treaties which are binding on the 

President if he ratifies the agreement. See id. at 96-100 (describing practice). Presumably, 

under the postulated response, the President could disregard the Senate's reservations 

even after he had ratified an agreement if the agreement had not in fact reached the treaty 

threshold. There are similar concerns over the vexed problem of treaty interpretation. 

The President may be limited in the extent to which he can reinterpret a treaty in a manner 

inconsistent with the understanding of the Senate when it gave its consent. See id. at 93-96 

(discussing treaty interpretation controversies). Would this limitation apply to an agreement 

consented to by the Senate but which did not really warrant the appellation "treaty" 

Apparently not, on the postulated view. Yet, all of these are important matters which 

would have surely drawn attention had anyone doubted that Senate approval ipso facto 

made an agreement an Article H treaty. 

360 Chief Justice Marshall noted that in construing the Constitution "the same words 

have not necessarily the same meaning attached to them, when found in different parts of 

the same instrument; their meaning is controlled by the context." Cherokee Nation v. 

Georgia, 30 U.S. (5 PeL) 1, 19 (1831).



[Vol. 73:1791 

compacts as an exception to the general prohibition on treaties, so 

that "treaties" represent the general category and "agreements" and 

"compacts" are a subset of the former. To make the point more perspicuously, 

it would be as if Section 10 provided: "No state shall enter 

into any treaty, alliance, or confederation; except they may, with the 

consent of Congress, enter into agreements or compacts. ' '361 In context, 

moreover, it is apparent that the Framers had compelling drafting 

reasons to adopt the existing formulation that are in no way 

incompatible with the view I have suggested. 362 

361 Some have thought that the Framers borrowed the distinction between treaties and 

agreements or compacts from Vattel. See, e.g., United States Steel Corp. v. Multistate Tax 

Comm'n, 434 U.S. 452, 462 n.12 (1978) (discussing Vattel's views in decision on Compact 

Clause); Abraham C. Weinfeld, What Did the Framers of the Federal Constitution Mean 

by "Agreements or Compacts", 3 U. Chi. L. Rev. 453, 457-60 (1936) (arguing that Framers 

had Vattel in mind in drafting Compact Clause). Vattel distinguished between "treaties" 

and "agreements, conventions, and arrangements": The former are perpetual or of 

considerable duration and call for "a continuous performance of acts," whereas the latter 

are perfected in a single act "once for all." 3 Vattel, supra note 257, bk. 2, ch. 12, §§ 152- 

153. But Vattel himself recognized that "agreements, conventions, and arrangements" 

(which Weinfeld identified with the Constitution's "agreements and compacts," see 

Weinfeld, supra, at 460) were simply forms of the general category "treaties." Thus, in a 

later section, he stated: 

Treaties which do not call for continuous acts, but are fulfilled by a single 

act, and are thus executed once for all, those treaties, unless indeed we prefer 

to give them another name (see § 153) [defining agreements, conventions, and 

arrangements], those conventions, those compacts, which are executed by an 

act done once for all .. .are, when once carried out, fully and definitely 

consummated. 

Id. bk. 2, ch. 12, § 192, at 172. It is worth noting that Vattel's distinction, even were it the 

source for the Framers' text, proved unworkable immediately and has never been reflected 

in actual practice. See 3 Story, supra note 202, § 1396 (rejecting George Tucker's earlier 

endorsement of (significantly) modified version of Vattel's distinction as "at best a very 

loose, and unsatisfactory exposition"). Those who have relied on it to distinguish between 

important and unimportant agreements have always been embarrassed by Vattel's failure 

to draw that as the point of difference. See Congressional Oversight of Executive Agreements, 

1975: Hearings on S. 632 and S. 1251 Before the Subcomm. on Separation of Powers 

of the Senate Comm. on the Judiciary, 94th Cong. 392 (1975) (reprinting memorandum 

by Monroe Leigh, Legal Adviser, Dep't of State, which noted that "Vattel's distinction has 

nothing to do with 'important' or 'unimportant"'). 

362 The Framers had important structural reasons to adopt their formulation. Section 10 

actually includes three paragraphs. See U.S. Const. art. I, § 10, cls. 1, 2, 3. The prohibition 

on treaty-making appears in the first along with a list of other powers denied the states; the 

permissive rule for agreements and compacts, in contrast, appears in the third along with 

other powers that can be exercised only with congressional consent. This grouping served 

to clarify and emphasize that certain powers (those listed in the first paragraph) were absolutely 

prohibited to the states but that others (those listed in the second and third) could be 

exercised with leave of Congress. See Holmes v. Jennison, 39 U.S. (14 Pet.) 540,571 (1840) 

(opinion of Taney, C.J.) (affirming distinction); 3 Story, supra note 202, § 1396, at 270 

(same); supra notes 218-19 and accompanying text (discussing same). Hence, the Framers 

were impelled to separate the treatment of state treaties and agreements by drafting imperatives 

that ruled out the more simple language I have suggested, but their drafting choice




In any case, even if we accept Professor Tribe's view that the term 

"treaty" in Section 10 does not comprehend "agreements or compacts," 

there are forceful textual reasons not to carry this limited construction 

over to the Treaty Clause. The term "treaty" in Section 10 is 

grouped together with the terms "alliance" and "confederation," and 

these are contrasted with "agreements" and "compacts." As Justice 

Story urged in his Commentaries, this grouping provides strong 

grounds for giving the term "treaty" a more limited sense in this context. 

Applying the maxim noscitur a sociis (the sense of each is best 

known by its association), he argued that the term treaty in Article I 

was limited to "treaties of a political character." 363 Compacts and 

agreements, in contrast, were best understood as applying to agreements 

dealing with "mere private rights of sovereignty. 

364 

It is hardly 

necessary to emphasize the difference in language between Articles I 

and II, the latter of which grants the treaty power to the President and 

the Senate in unrestricted terms and without making reference to 

agreements, compacts, alliances, or confederations. Context, then, 

strongly suggests that the limited meaning arguably appropriate in Article 

I has no application in Article fl.365 

This is particularly so when one considers the consequences-for 

the whole point of Professor Tribe's argument is that his construction 

opens up a hole in the Constitution's grant of the foreign affairs powers 

to the national government. It is hardly sensible to have recourse 

provides no compelling reason for doubting that my construction accurately corresponds 

with their intent 

363 3 Story, supra note 202, § 1397, at 271. 

364 3 id. § 1397, at 272. These dealt with "questions of boundary; interests in land, situate 

in the territory of each other, and other internal regulations for the mutual comfort, 

and convenience of states, bordering on each other." Id. Indeed, because of his sense that 

the term "treaty" was properly a general category, Justice Story was tempted to conjecture 

that "the original reading was 'treaties of alliance, or confederation.'" Id. at 271 n.2. He 

quite properly recognized, however, that this specific explanation was ruled out by the 

drafting history. See id. (concluding that corresponding provision of articles of confederation 

undermines this conjecture). 

365 Nor is Chief Justice Taney's opinion in Holmes to the contrary; indeed, it supports 

the traditional view. Taney emphasized the parallel between the use of the term "treaty" 

in Articles I and II, not to demonstrate that the term had a limited meaning in Article II, 

but to emphasize the comprehensiveness with which states were denied independent 

agreement-making powers. See Holmes, 39 U.S. (14 Pet.) at 571-72 (opinion of Taney, 

CJ.). Indeed, he repeatedly stressed the sweeping scope of the treaty power in Article II, 

noting that it "is given by the Constitution in general terms, without any description of the 

objects intended to be embraced by it; and, consequently, it was designed to include all 

those subjects, which in the ordinary intercourse of nations had usually been made subjects 

of negotiation and treaty." Id. at 569. He also repeatedly emphasized that the power to 

make extradition agreements was both within the treaty power and could properly be the 

subject of a state agreement or compact if approved by Congress under the Compact 

Clause. See id. at 568-69, 573-76. He must, then, have rejected Professor Tribe's view.



[Vol. 73:1791 

to extraordinary theories of unenumerated powers to ffll in a gap in 

constitutional space when there are perfectly plausible, even compelling 

interpretive alternatives that obviate the problem. It is hardly 

sensible, that is, unless one is a priori committed to upholding the exclusivity 

of the Treaty Clause and is willing to bend, stretch, and perhaps 

even tear the text in order to reach a preordained result. 

Perhaps, then, a case of bad conscience explains Professor Tribe's 

otherwise bizarre concession, hidden in a footnote in an unrelated section 

of his argument: "Of course, the Treaty Clause of Article II 

would also seem to make a 'treaty' out of any agreement approved in 

accordance with the Treaty Clause's terms, even if the President 

would not have actually needed Senate supermajority approval of the 

agreement. '366 "Of course." But then what happened to the troublesome 

gap that supposedly motivates the whole argument 

But let us put aside this devastating footnote concession, and proceed 

as if Professor Tribe simply had not contradicted his own position-because 

the textual difficulties with his view do not end here. 

His peripheral vision, it seems, is badly impaired. Thus, for example, 

if we take his formalistic interpretive approach seriously, the difficulties 

he purports to uncover run far deeper than he imagines. Recall 

that Article I, Section 10 distinguishes not only between "treaties" and 

"agreements or compacts," but between "treaties," "alliances," and 

"confederations" as well. Presumably, the Framers did not just throw 

in these additional terms to lengthen the document; in their usage, if 

we follow Professor Tribe, "treaties" must not include "alliances" or 

"confederations." Yet, any careful reader ought immediately to notice 

that the Framers conspicuously failed to include the latter terms in 

Article II. Are we then to conclude that the President and the Senate 

can make "treaties" but not "alliances" and "confederations" But if 

the exclusion of "agreements" from the treaty power opened up a 

hole in constitutional space, the exclusion of "alliances" and "confederations" 

would seem to open a giant chasm. And if the power does 

not belong to the President and the Senate, and if the Congress as a 

whole is excluded, are we to conclude that the power falls within the 

President's unilateral authority as chief executive Is our participation 

in the NATO "alliance," for example, within the President's unilateral 

powers Is NAFTA a "confederation" The implications 

would be undeniably mind-boggling, if they were not so obviously 

silly. 

Professor Tribe's peripheral vision fails in another crucial respect 

as well. He somehow misses the fact that the term "treaties" appears 

366 Tribe, Taking Text, supra note 3, at 1272 n.175.




not only in Articles I and II but in Articles Ill and VI as well. Recall 

that the judicial power extends to all cases arising under the Constitution, 

laws, and "treaties" of the United States a67 and that the 

Supremacy Clause likewise declares that the Constitution, the laws, 

and "treaties" are the supreme law of the land. 36s To be sure, these 

clauses appear at first glance to be simple cognate provisions to Article 

I's Treaty Clause-any "treaty" to which the Senate gives its consent 

under Article II falls within the judicial power of the national 

courts and is supreme law of the land. Yet, they have nevertheless 

been given a more expansive construction. In B. Altman & Co. v. 

United States, 369 the Supreme Court held that the term "treaty" in the 

Judiciary Act of 1891 included not only treaties approved by the Senate 

but executive agreements concluded by the President as well.37 0 

Since then most have assumed that Article III's reference to "treaties" 

includes all international agreements to which the United States is a 

party, regardless of the method through which they were processed 3 71 

Likewise, it has often been assumed that executive agreements are 

supreme law of the land because they are "treaties" for purposes of 

Article VI.372 The reason is straightforward: A broad construction is 

necessary in order to avoid just what Professor Tribe seeks to exploit, 

a hole in constitutional space. Cases arising under executive agreements 

would otherwise be relegated to the state courts, and state laws 

could thwart the foreign policies of the President. These are precisely 

the kinds of gaps a good faith interpreter ought to avoid whenever 

possible. If, then, the term "treaty" in Articles Ill and VI includes all 

agreements the United States enters no matter how approved-in direct 

contrast to the more limited construction Professor Tribe gives it 

in Article I-then surely its use in Article II is expansive enough to 

367 See U.S. Const. art. HI, § 2, cl. 1. 

368 See id. art. VI, cl. 2. 

369 224 U.S. 583 (1912). 

370 Id. at 600-01. 

371 See Restatement, supra note 9, § 111 cmt. e & reporters' note 4 (adopting this view); 

cf. Weinberger v. Rossi, 456 U.S. 25, 28-31 (1982) (construing term "treaty" in statute 

prohibiting discrimination unless permitted by treaty to include congressional-executive 

agreements). 

372 See Restatement, supra note 9, § 111 Cmt. d & reporters' note 2 (suggesting this 

view). In United States v. Belmont, 301 U.S. 324,331 (1937), Justice Sutherland held that 

sole executive agreements are supreme over state law but did not expressly rest on language 

in the Supremacy Clause. But, as the Restatement recognizes, "[t]he same result 

might be reached under the Supremacy Clause by giving the same broad interpretation to 

the words 'treaties' and 'laws' that has been adopted for purposes of judicial power and 

jurisdiction." Restatement, supra note 9, § 111 reporters' note 2.



[Vol. 73:1791 

include any international agreement actually given the Senate's advice 

and consent. 373 

Finally, Professor Tribe's position suffers from still another serious 

structural defect. By finding presidential authority implicit in the 

Compact Clause, he is forced to the anomalous view that the unilateral 

powers of the President are closely parallel to the power of the 

states to enter into agreements or compacts. He seems to think that 

hard-nosed textual interpretation requires that they be limited by a 

common metric. 374 This position, however, is utterly implausible. 

Although we have very little precedent on which to base an authoritative 

judgment, it has been widely assumed from early on that the 

Compact Clause could have been meant to open only a narrow space 

for the states to deal with local matters through foreign negotiations. 

375 In contrast, the President is concerned principally with matters 

of national and international importance. He makes armistices, 

373 If, alternatively, one were to take the view that executive agreements are within the 

judicial power and are supreme law of the land not because they are "treaties," but because 

they are "laws," it would do more harm than benefit to Professor Tribe's position. 

This view would remove one unnecessary argument against his interpretation of the term 

"treaty" but would simultaneously seriously undermine his Necessary and Proper Clause 

argument. If executive agreements are "laws" under the Supremacy Clause, then surely a 

resolution approving an executive agreement would be a "law" under the Necessary and 

Proper Clause. See supra Part III.B.1; supra note 243. This is not to suggest that Professor 

Tribe's Necessary and Proper Clause argument is, in any case, plausible. On the other 

hand, one might join with Justice Sutherland and suppose that the foreign affairs powers 

are extratextual and that unilateral agreements are supreme law of the land and within the 

federal judicial power simply because of the nation's status as a sovereign under international 

law. However, it is difficult to imagine that Professor Tribe would wish to opt for 

this path, while simultaneously defending his strictly textualist approach to the Teaty 

Clause. If the text can be radically supplemented in this way, there would be little reason-or 

justification-for standing on textual formalities in opposing Congress's powers to 

approve agreements. 

374 See Tribe, Taking Text, supra note 3, at 1266 & n.154. 

375 See, e.g., Henkin, supra note 9, at 152-53 (affirming this view); 3 Story, supra note 

202, § 1397 (same). In fact, there have been very few actual state agreements with foreign 

nations, and these have been confined to matters of local interest, mostly necessitated by 

common borders. See Henkin, supra note 9, at 153 (describing some of agreements approved 

by Congress). Chief Justice Taney's opinion in Holmes v. Jennison, 39 U.S. (14 

Pet.) 540 (1840), is emphatic in suggesting the limited role of the states in foreign negotiations. 

See id. at 572-75 (opinion of Taney, C.J.) (declaring that state involvement would be 

"totally contradictory and repugnant" to constitutional plan). Some have thought, however, 

that the validity of state agreements is controlled by purely political considerations 

within the discretion of Congress and that the courts are to stand aloof. See, e.g., Henkin, 

supra note 9, at 153 (noting that "[i]t would be difficult to believe .. the courts would 

invalidate... [a state] agreement to which Congress consented"). The classic argument to 

this effect, although focusing on interstate compacts, rather than compacts with foreign 

states, is in Felix Frankfurter & James M. Landis, The Compact Clause of the Constitution-A 

Study in Interstate Adjustments, 34 Yale L.J. 685, 707-08 (1925) ("[M]ost questions 

of interstate concern are beyond the jurisdiction of the Supreme Court; they are 

beyond all court relief. Legislation is the answer .... ).




concludes agreements incident to the recognition of foreign states, settles 

claims, and enters agreements repairing relations with adversaries 

through a myriad of mechanisms. The subjects appropriate for treatment 

by the states and the President could not be more different. 

Thus, the notion that the two powers derive from the same source and 

are roughly comparable in scope is unsustainable. 37 6 

It is worth noting that despite the Compact Clause's unconditional requirement of 

congressional consent for state agreements and compacts, the Supreme Court has permitted 

states to enter agreements without congressional approval so long as these do not tend 

"to the increase of political power in the States, which may encroach upon or interfere with 

the just supremacy of the United States." Virginia v. Tennessee, 148 U.S. 503, 519 (1893); 

see also United States Steel Corp. v. Multistate Tax Comm'n, 434 U.S. 452, 468-71 (19r78) 

(reaffirming this view). This should be another warning against rigid interpretive presumptions 

of the kind on which Professor Tribe relies. 

376 Holmes at least implicitly supports this conclusion, though in a way that underscores 

the narrowness of the early understandings of the President's unilateral authority. Thus, in 

his opinion, Chief Justice Taney repeatedly affirmed the traditional understanding that the 

President has no power to extradite a person to a foreign nation in the absence of a treaty 

or an act of Congress. See Holmes, 39 U.S. (14 Pet.) at 574 (opinion of Taney, CJ.). But 

he equally affirmed the power of the states to do so by compact if they obtain the consent 

of Congress. See id. at 578-79. He thus saw no parallel. 

Still another objection to Professor Tribe's view is implicit in what has already been 

said: His effort to locate the President's unilateral authority in an unenumerated power to 

make agreements simply fails the tests of history and of text. The President enters into 

unilateral agreements, it has been widely understood, not from a general agreementmaking 

authority, but as an incident of his substantive powers. See supra note 329 and 

accompanying text. Thus, contrary to Professor Tribe's assertion, we cannot determine the 

validity of a presidential agreement solely by reference to the level of its impact on state or 

national sovereignty (on the supposition that this reflects the difference between "agreements" 

and "treaties"). We must first determine whether the agreement is sufficiently related 

to an aim within the President's substantive powers. During the first 150 years of our 

history, those powers were understood far more narrowly than today- as a consequence, 

the scope of the President's unilateral powers was extremely modest. Curtiss-Wright, 

Belmont, and Pink, however, mark the beginnings of an enormous expansion in the conception 

of independent presidential authority in foreign affairs. See United States v. Pink, 

315 U.S. 203, 228-30 (1942) (upholding Litvinov Assignment); United States v. Belmont, 

301 U.S. 324,330-31 (1937) (same); United States v. Curtiss-Wright xport Corp., 299 U.S. 

304, 318-21 (1936) (reciting expansive dicta on President's foreign affairs powers). The 

resulting dilemma is stark: Unless we are prepared to render the Treaty Clause a virtual 

nullity, there must be limits on the President's sole authority, but, given his plenary powers, 

how to derive those limits is not clear. See Henkin, supra note 9, at 222 (observing plaintively 

that "[o]ne is compelled to conclude that there are agreements which the President 

can make on his sole authority and others which he can make only with the consent of the 

Senate (or of both houses), but neither Justice Sutherland nor any one else has told us 

which are which" (footnotes omitted)). Ironically, then, in order to preserve the core value 

expressed in the Treaty Clause, we may have no choice but to fall back on a Tribe-like 

notion that there are some agreements that are simply too important to be concluded by 

the President alone. (This is not to suggest, though, that Professor Tribe is the first commentator 

to suggest the idea. See, e.g., Edwin Borchard, Shall the Executive Agreement 

Replace the Treaty, 53 Yale LJ. 664, 669-70 (1944) (arguing that executive agreements 

are appropriate only for routine or unimportant matters).) Our interpretive options would



[Vol. 73:1791 

We thus have an impressive array of reasons to put aside Professor 

Tribe's approach. Perhaps, however, there are other arguments, 

which Professor Tribe has simply missed or foregone, that might more 

successfully distinguish between the unilateral and the congressionalexecutive 

agreement. At least one other, also based on the Compact 

Clause, ought to be considered, since it has sometimes been put forward 

in explanation of the unilateral agreement and could potentially 

aid Professor Tribe's position. 377 It has the additional merit of not 

being based on the false premise which mars his own approach. 

be much broader, of course, were we to reject Professor Tribe's textualism and look to 

history for appropriate limits on the President's unilateral powers. 

Finally, although I do not have space to consider it here, Professor Tribe's effort to 

define the dividing line between agreements and treaties is unsatisfactory. See Tribe, Taking 

Text, supra note 3, at 1266-68 & 1267 n.156 (suggesting that treaty status is ultimately 

determined by agreement's impact on state or national sovereignty, and arguing that burden 

WTO Agreement's dispute resolution mechanism imposes on state sovereignty renders 

it a treaty). Most importantly, his focus on dispute resolution procedures is 

misguided. Not only were these often included in unilateral agreements, even in the nineteenth 

century, see Moore, supra note 20, at 408-17 (describing nineteenth century claims 

settlements including arbitration provisions), there is nothing inherent in the inclusion of 

such a provision that renders an agreement ipso facto beyond the President's unilateral 

authority. That view overlooks the entire corpus of the international law of treaties and of 

state responsibility, which not only makes treaties binding but provides remedies for their 

breach. See, e.g., Vienna Convention on the Law of Treaties, opened for signature May 23, 

1969, arts. 60, 65-66, 1155 U.N.T.S. 331, 346-48 (providing, inter alia, for suspension or 

termination of treaties in response to material breach); Draft Articles on State Responsibility, 

arts. 41-46, Report of International Law Commission, U.N. GAOR, 51st Sess., Supp. 

No. 10, at 125, 141-43, U.N. Doc. A/51/10 (1996) (defining rights of states which are victims 

of internationally wrongful act), reprinted in 37 I.L.M. 440, 454-56 (1998). Dispute resolution 

procedures are not necessarily a greater restriction on domestic sovereignty; they simply 

provide a fair and impartial method for resolving conflicts that might otherwise be 

resolved by the force of power rather than the force of the better argument. In any case, 

just as we have the power to violate a treaty stipulation, we also have the power to disregard 

the rulings of an international tribunal and, unfortunately, have not always been shy 

about exercising that power. For an extended consideration of the United States's withdrawal 

from Nicaragua's suit against it in the International Court of Justice, see the articles 

collected in The International Court of Justice at a Crossroads (Lori Fisler Damrosch ed., 

1987). 

Professor Tribe complains that Professor Ackerman and I have disparaged as ad hoc 

the effort to draw lines between treaties and executive agreements. See Tribe, Taking Text, 

supra note 3, at 1266. In fact, what we "disparaged" was his effort to validate most 

congressional-executive agreements but deny the constitutionality of those dealing with 

particularly important matters. See Ackerman & Golove, supra note 5, at 921-22. He 

apparently now agrees, having reversed views and written off all congressional-executive 

agreements. It is of course true that drawing a line between agreements that may be concluded 

on the President's sole authority and those which can only be concluded as treaties 

or congressional-executive agreements is extremely difficult, but that difficulty does not 

excuse a failure to do our best. 

377 See, e.g., Borchard, supra note 376, at 667-70 (articulating argument based on Compact 

Clause); Richard Cohen, Self-Executing Executive Agreements: A Separation of 

Powers Problem, 24 Buff. L. Rev. 137, 139-40 (1974) (same).




The argument runs as follows: By permitting states to make 

agreements or compacts with congressional consent, the Framers revealed 

that they did not deem the full protection of Senate 

supermajority consent vital for this lesser category of international undertakings. 

It is admittedly odd that they did not expressly provide an 

alternative procedure for federal agreements and compacts. Nevertheless, 

the inference is strong, or so it is claimed, that they did not 

believe that these lesser undertakings, whether made by the states or 

the federal government, require the heightened scrutiny of senatorial 

consent. Why would they have allowed the states this option but have 

denied it to the national government There are grounds, then, for 

interpreting the Treaty Clause to permit the approval of "agreements" 

through a less burdensome procedure, and the power thus revealed 

belongs to the President. 

With the exception of the hole in constitutional space argument, 

it should be clear that this view, which is really only a modified version 

of Professor Tribe's position, is subject to all of the same objections 

applicable to his original argument. 378 But there is a further 

378 This version presumes, for example, that the Executive Power Clause is a general 

grant of all powers that are executive in character, that there is an undifferentiated presidential 

agreement-making power, that the powers of the states and the President derive 

from the same source and are roughly parallel, and that agreement-making is an executive 

function. The flaws in these arguments have already been demonstrated. As to the last, it 

may be added that it is extremely difficult to see why we should disregard the Framers' 

express preference for congressional supervision over agreement-making in the state context 

when moving into the non-text-based, parallel federal agreement-making power. This, 

of course, seriously undermines the argument for executive unilateralism. Even were we 

to avoid this problem by recognizing a congressional rather than presidential power to 

approve "agreements," however, this would simply introduce a new difficulty. It would 

necessarily mean that approving some agreements falls within the Necessary and Proper 

Clause. In that case, Professor Tribe's claim that the term "laws" does not include resolutions 

approving international agreements, whatever force it would otherwise have had, 

could not be sustained. See supra Part III.B.1; supra note 243. Moreover, if the Compact 

Clause suggests that the Framers must have assumed that Congress could approve presidential 

"agreements," then their failure to specify such a power in Article I, Section 8 

suggests, in turn, that they assumed it was already within the Necessary and Proper Clause. 

That clause, however, does not distinguish between "agreements" and "treaties" and 

thereby creates a strong inference that the distinction does not apply to Congress's implied 

powers to approve federal commitments. It would be an interpretive leap of faith to find 

implicit limits on Congress's implied powers in a provision affirmatively giving Congress 

supervisory authority over state agreement-making. Thus, the more persuasive version of 

the argument from the Compact Clause leads right back to the modem congressionalexecutive 

agreement. 

Professor Tribe professes to find this argument "astonishing." Tribe, Taking Text, 

supra note 3, at 1272. In particular, he takes us to task for concluding that "we do not see 

how the words of the Compact Clause-designed for a very different problem-are relevant 

in determining the scope of congressional power under Article I[, Section 8]." 

Ackerman & Golove, supra note 5, at 921 n.514 (emphasis added), quoted in Tribe, Taking 

Text, supra note 3, at 1271. Once again, of course, we made this point in response to


1918 


[Vol. 73:1791 

objection to this specific form of the argument that is of greater importance 

for our purposes. By eschewing reliance on the false underpinnings 

of Professor Tribe's view, this version is forced to concede 

the incorrectness of his basic premise about the exclusivity of the 

Treaty Clause. Since there is no gap, the President may admittedly 

conclude unilaterally what he and the Senate are empowered to conclude 

through the advice and consent procedure. Yet, if the arguments 

for the Treaty Clause's exclusivity give way in the face of the 

rather remote string of inferences from the Compact Clause, then it is 

exceedingly difficult to see why they do not collapse in the face of the 

much stronger and more direct arguments in favor of the power of the 

national legislature. The Compact Clause, after all, appears in a section 

devoted to limits on the powers of the states and does not even 

mention federal agreements, let alone unilateral presidential authority. 

379 Congress, in contrast, is explicitly vested with supervisory authority 

not only over state agreements but over virtually the whole of 

our foreign affairs. The Compact Clause argument, in short, has no 

way of explaining why we should accept presidential but not 

congressional-executive agreements. 380 

Professor Tribe's then view that some agreements could be approved by Congress but 

others were beyond the scope of its authority. Although he finds the argument "astonishing," 

he seems to have drastically revised his view just to avoid its force. What else can 

explain the vigor with which he denies Congress any authority to approve presidential 

agreements, even those which it could approve if negotiated by the states-even though he 

concedes that under his reading the Framers' scheme presents "a puzzle." Tribe, Taking 

Text, supra note 3, at 1271 n.172. He never seems to contemplate the possibility that this 

puzzle provides a reason for further reflection on whether the interpretation he insists 

upon might require revision, or at least be less compelling than he had at first thought. 

379 Indeed, the most persuasive inference as to why the Framers assigned supervision of 

state agreements to the Congress rather than to two-thirds of the Senate is not that they 

wished to provide a less burdensome procedure for minor agreements. On the contrary, it 

is that they thought the House's participation in this class of essentially local agreements 

did not pose the same threat to the interests in secrecy, dispatch, and long-term perspective 

that might apply to agreements affecting the nation as a whole. When these considerations 

did not require otherwise, the Framers were content to let the national legislature play its 

ordinary role. This, in turn, suggests that they had no aversion to House participation in 

approving agreements so long as the threats about which they were concerned were not 

present. And this is precisely what the congressional-executive agreement permits: When 

the President does not believe that House participation threatens these important interests, 

he may seek the approval of Congress as a whole rather than the approval of one House by 

supermajority vote. For further discussion, see supra notes 68-75, 186, and accompanying 

text. Furthermore, the fact that the Framers assigned the supervisory role to Congress 

rather than the President is yet further evidence that they viewed approving agreements as 

a legislative not an executive task. See supra notes 247-62 and accompanying text. 

380 Professor Tribe might also have tried a functionalist line of argument. He might 

have conceded, for example, that the Treaty Clause cannot be exclusive and that the unilateral 

executive agreement is justifiable only as an implied power of the President incident to 

his executive authority. He might have further conceded that presidential agreements are 

a derogation from the protections afforded state interests by the Treaty Clause. Still, citing



1919 

E. A Closing Flourish: Competing for the Father's Approval- 

Who is Really Marshallian 

Professor Tribe takes one final crack at the expressio unius line of 

attack, although it is not entirely clear whether he means his argument 

to be rhetorical or substantive. Taking umbrage at Professor 

Ackerman and me for dubbing our interpretation of the Necessary 

and Proper Clause "Marshallian," he wishes to make clear that it is 

anything but and that "[t]he great Chief Justice would likely distance 

himself" from arguments like ours. 3 81 Indeed, he points out, "we need 

look no further than Chief Justice Marshall's opinion in Marbury v. 

Madison" to discover his attachment to the expressio unius principle. 

3 8 Marshall's holding, after all, was that Article IM's "architectural" 

3 8 specification of cases within the Supreme Court's original 

jurisdiction is exclusive notwithstanding the absence of the word 

"only.", 

It is, of course, true that Chief Justice Marshall held that Congress 

could not add to the Court's original jurisdiction and, in doing 

so, implicitly invoked the expressio unius canon. 3 8 But it is instructhe 

sole organ power, he could have claimed that they are justifiable because the Constitution 

vests control over the conduct of foreign affairs in the President. In the external 

realm, he might have argued, were the President not accorded some flexibility in concluding 

agreements unilaterally, our foreign relations would be hamstrung. The President is on 

the front-line, but Congress, as important as its foreign affairs powers are, has a mostly 

supervisory role. Thus, according to this view, its powers do not require the kind of flexibility 

essential to executive efforts, nor justify the significant infringement of federalism 

interests that recognizing the congressional-executive agreement would require. 

To be sure, given the historical evidence about the Framers' actual purposes in adopting 

the Treaty Clause, this argument may give a reasonable account of what they might 

have had in mind. However, it is critical to recognize that its plausibility, if any, is almost 

entirely parasitic on history. As a matter of purely textual exegesis, it rests on extravagant 

inferences from the President's already textually problematic sole organ power. See supra 

notes 309-18 and accompanying text. Moreover, it overrides both the strong expressio 

unius inference and the federalism arguments against presidential unilateralism, see supra 

notes 326-29 and accompanying text, on the basis of speculative functionalist considerations 

that the Framers, for whatever reasons, did not incorporate into the text. In this 

respect, it is even weaker than the Compact Clause argument. Given the lack of textual 

support for this view, then, it does little to advance Professor Tribe's textualist agenda. 

This just drives home the main point: The compulsion to find a tertual basis to explain the 

unilateral agreement while ruling out the congressional-executive agreement is just that-a 

drive that is motivated not by textual imperatives but by a prior commitment to reaching a 

certain result In any case, even were this functionalist account more textually compelling 

than I believe, it still would not aid Professor Tribe. At best, it would only demonstrate 

that the text is, as I have said all along, indeterminate. 


382 Id. 

383 Id. 

384 See Marbury v. Madison, 5 U.S. (1 Cranch) 137, 174-75 (1803). The relevant clause 

of Article II provides:



[Vol. 73:1791 

tive to look more precisely at what he said. For rather than blunderbuss 

assertion of the maxim, he was at pains to show that anything but 

an exclusive construction would render the provision "mere surplusage,... 

entirely without meaning,... form without substance. 38 5 His 

discussion is thus consistent with Hamilton's in Federalist No. 83 and 

with the traditional authorities I cited earlier. The expressio inference 

is most persuasive when necessary to avoid draining a provision of all 

meaning and purpose. When not necessary to that end, its force is 

greatly reduced. 38 6 As we have seen, accepting the congressionalexecutive 

agreement, far from rendering the Treaty Clause a nullity, 

leaves it substantial scope. 38 7 

However, the inapplicability of Marshall's reasoning is hardly the 

main point. Is there self-conscious irony in Professor Tribe's valorization 

of Marshall's performance in Marbury as an exemplar of original 

meaning jurisprudence Marbury was certainly as political a decision 

as any in the history of the Court, its outcome preordained by the 

overwhelming political power of the Republican Congress and White 

House arrayed against the Federalist Court. 388 The Chief Justice's 

opinion is justly celebrated not for the rigor of its legal argumentation 

but for its remarkable political savvy. Under the most inauspicious 

circumstances, Marshall was able to lay the foundations for an expansive 

conception of the judiciary's role while strategically conceding the 

In all Cases affecting Ambassadors, other public Ministers and Consuls, and 

those in which a State shall be Party, the supreme Court shall have original 

Jurisdiction. In all the other Cases before mentioned, the supreme Court shall 

have appellate Jurisdiction, both as to Law and Fact, with such Exceptions, and 

under such Regulations as the Congress shall make. 

U.S. Const. art. III, § 2, cl. 2. 

385 Marbury, 5 U.S. (1 Cranch) at 174. 

386 See supra notes 83-87, 92-94, and accompanying text. 


388 As the last Federalist stronghold after the Jeffersonian sweep in 1800, the Court was 

under intense political attack, and its institutional position was tenuous at best. By repealing 

the Judiciary Act of 1801, Congress had already unseated 16 sitting federal circuit court 

judges, and impeachment proceedings against another were under way. Congress had even 

delayed the Supreme Court's term for over a year in response to the Court's temerity in 

issuing an order to show cause to Secretary of State Madison. Under the circumstances, 

the Court had no choice but dismiss the suit or watch its order go unenforced and its 

institutional position crushed. For an engaging historical account emphasizing Marshall's 

role in the intense political controversies of the moment, see Bruce Ackerman, The Roots 

of Presidentialism 97-140 (1997) (unpublished manuscript, on file with the New York University 

Law Review); see also Dean Alfange, Jr., Marbury v. Madison and Original Understandings 

of Judicial Review: In Defense of Traditional Wisdom, 1993 Sup. Ct. Rev. 329, 

349-72 (providing detailed historical account); James M. O'Fallon, Marbury, 44 Stan. L. 

Rev. 219 (1992) (same). The parallels to the Court's position at the height of the New Deal 

crisis should be clear.




case to Jefferson 3 89 Brilliant though his effort might have been, however, 

the opinion's weaknesses as a "legal" text are barely submerged 

below the surface of his masterful rhetoric. 

Begin with the fact that Marshall was himself a party to the dispute 

and an important witness. It was Marshall, after all, who as Secretary 

of State had negligently failed to deliver the sealed commission 

to Marbury before inauguration day; he was also a key witness to a 

crucial fact, that the commission had actually been signed and 

sealed. 390 One can hardly imagine a clearer case for recusal. Second, 

while Marshall famously dismissed Marbury's claim for want of jurisdiction, 

the bulk of the opinion is a lengthy disquisition on the merits 

self-evidently designed to embarrass the new chief executive and provide 

political hay for the Federalists. 39 ' By what right did the Chief 

Justice gratuitously consider and decide the merits, only to dismiss for 

lack of jurisdiction 

But it is not only that the opinion was obviously politically motivated. 

Marshall had to strain to find a constitutional issue and then 

strain again to find Section 13 of the Judiciary Act of 1789 unconstitutional. 

There is a virtual consensus that he could easily have avoided 

the constitutional conflict simply by construing Section 13 not to provide 

original jurisdiction over mandamus actions. Most have thought 

that this was the better interpretation in any event. 39 2 More to the 

point, the widely accepted historical verdict is that Marshall's reasoning 

on the constitutional question was defective. Indeed, expressio 

unius notwithstanding, most modem scholars think he got it wrong: 

389 These features of the decision have been frequently remarked upon. See, e.g., 

Robert G. McCloskey, The American Supreme Court 25-28 (Sanford Levinson ed., 2d ed. 

1994); Alfange, supra note 388, at 366-72, 380-83; William NV. Van Alstyne, A Critical 

Guide to Marbury v. Madison, 1969 Duke LJ. 1, 34-38. 

390 See Alfange, supra note 388, at 392-93 (describing Marshall's role); Ackerman, supra 

note 388, at 74-75 (same); Van Astyne, supra note 389, at 8 (same). 

391 See Alfange, supra note 388, at 387-91 (discussing Marshall's agenda); Edward S. 

Corwin, Marbury v. Madison and the Doctrine of Judicial Review, 12 Mich. L Rev. 538, 

542-43 (1914) (same); Van Alstyne, supra note 389, at 6-8 (same). As Corwin characteristically 

put it: 

To speak quite frankly, this decision bears many of the earmarks of a deliberate 

partisan coup. The court was bent on reading the President a lecture on his 

legal and moral duty to recent Federalist appointees to judicial office... but at 

the same time hesitated to invite a snub by actually asserting jurisdiction of the 

matter. 

Corwin, supra, at 542-43. 

392 See, e.g., Corwin, supra note 391, at 541-43 (arguing on this ground that Marshall 

misinterpreted Section 13); David P. Currie, The Constitution in the Supreme Court: The 

Powers of the Federal Courts, 1801-1835, 49 U. Cii. L Rev. 646, 653 (1982) (same); Van 

Aistyne, supra note 389, at 14-16 (same). Although Marshall read Section 13 to grant the 

Court original jurisdiction over mandamus actions, that section seems only to have authorized 

the Court to issue writs in cases otherwise properly within its jurisdiction.



[Vol. 73:1791 

Article III should have been read to permit Congress to add to the 

Court's original jurisdiction.393 That was the contemporaneous view 

of the First Congress and the implicit view of the Supreme Court in 

earlier cases. 394 Most importantly, though, Marshall's claim that a 

nonexclusive reading would render the provision surplusage was extremely 

dubious. 395 This would seem to make Marbury an extremely 

unappealing precedent for Professor Tribe's original meaning 

argument. 

The irony, however, goes deeper. Professor Tribe thinks "[i]t 

cannot escape notice" that Marshall read the Original Jurisdiction 

Clause as exclusive. 396 However, it also ought not escape notice that 

the Chief Justice himself later repudiated the heart of this argument. 

393 See, e.g., Alfange, supra note 388, at 397-405 (arguing for nonexclusive reading of 

Original Jurisdiction Clause); Corwin, supra note 391, at 539-41 (same); Currie, supra note 

392, at 653-55 (same); Van Alstyne, supra note 389, at 30-33 (same). Even Beveridge, 

Marshall's partisan biographer, viewed his reasoning as "a pretext" for establishing the 

power of judicial review, 3 Albert J. Beveridge, The Life of John Marshall 133 (1916), and 

considered Marshall's constitutional argumentation, in the face of the contrary view of the 

First Congress, to be, charitably, the "only original idea" in the opinion, id. at 128. "Nobody," 

Beveridge claimed, "ever had questioned the validity of that section," which had 

been written by "[Oliver] Ellsworth, who preceded Marshall as Chief Justice" and who 

"was one of the greatest lawyers of his time and an influential member of the Constitutional 

Convention." Id. Notwithstanding, there have been dissenting voices. See, e.g., 

Robert Lowry Clinton, Marbury v. Madison and Judicial Review 96-97 (1989) (defending 

version of Marshall's rationale); Akhil Reed Amar, Marbury, Section 13, and the Original 

Jurisdiction of the Supreme Court, 56 U. Chi. L. Rev. 443, 463-78 (1989) (declining to rest 

on Marshall's reasoning but affirming his result). 

394 See Alfange, supra note 388, at 403-05 (discussing relevant history); Susan Low 

Bloch & Maeva Marcus, John Marshall's Selective Use of History in Marbury v. Madison, 

1986 Wis. L. Rev. 301, 326-33 (same); O'Fallon, supra note 388, at 256 (same). Bloch and 

Marcus in particular demonstrate how shoddily Marshall dealt with inconsistent 

precedents. 

395 See Alfange, supra note 388, at 399-400 (criticizing Marshall's argument); Corwin, 

supra note 391, at 540 (same); Currie, supra note 392, at 654-55 (same); Van Alstyne, supra 

note 389, at 31-32 (same). As these scholars have pointed out, contrary to Marshall's rhetoric, 

allowing Congress to add to the Court's original jurisdiction would not have left the 

Original Jurisdiction Clause without operation. It could be read to specify a minimum but 

not a maximum, so that Congress could add to but not subtract from the constitutionally 

assigned categories of cases. Alternatively, the division between original and appellate 

jurisdiction could have been a default assignment subject to congressional revision. The 

particular wording of the sentence that specifies the Court's appellate jurisdiction poses 

some difficulties. See U.S. Const. art. III, § 2, cl. 2. Marshall read the affirmative reference 

to appellate jurisdiction as an implicit negating of the power to add to original jurisdiction. 

Here, again, though, alternative readings were available. See Alfange, supra note 388, at 

398-99 (arguing that affirmative reference to appellate jurisdiction was made to provide 

context for granting Congress power to make exceptions to Court's appellate jurisdiction); 

Corwin, supra note 391, at 540-41 (arguing that Marshall need not have construed affirmative 

reference as limited to appellate jurisdiction, i.e., that Court's jurisdiction in these 

cases could be both appellate and original). 




AGAINST FREE-FORM FORALISM 

1923 

In Marbury, applying expressio unius, he had written: "If congress 

remains at liberty to give this court appellate jurisdiction, where the 

constitution has declared their jurisdiction shall be original; and original 

jurisdiction where the constitution has declared it shall be appellate; 

the distribution of jurisdiction, made in the constitution, is form 

without substance." 3 ' 97 But in Cohens v. Virginia, 393 he reversed direction, 

deriding this language as mere dicta. 399 Although he stubbornly 

clung to the view that Congress could not add to the Court's original 

jurisdiction, he now disowned the implications of his own earlier argument. 

Congress could not give original jurisdiction where the Constitution 

gave appellate, but it could grant the Court appellate 

jurisdiction where the Constitution gave original. 400 Without acknowledging 

the tension this created with his earlier reasoning, he 

simply reiterated that in Marbury the provision had to be given an 

exclusive sense to avoid rendering it "totally inoperative. '401 But, he 

added: 

The effort now made is, to apply the conclusion to which the court 

was conducted by that reasoning, in the particular case, to one in 

which the words have their full operation, when understood affirmatively, 

and in which the negative or exclusive sense, is to be so used 

as to defeat some of the great objects of the article. 402 

In other words, Marshall disowned the very sort of rigid expressio 

unius argument Professor Tribe puts forward as decisive. According 

to the "great Chief Justice," where an exclusive reading is unnecessary 

to avoid rendering a provision nugatory, the maxim should not be applied 

in the face of more persuasive arguments to the contrary. 403 

397 Marbury v. Madison, 5 U.S. (1 Cranch) 137,174 (1803); see also id. at 175 (repeating 

same). 

398 19 U.S. (6 Wheat.) 264 (1821). 

399 See id. at 399-401. The need to disregard Marbury prompted Marshall to give a 

classic statement of the proper weight to be accorded dicta. See id. at 399-400. 

409 See id. at 395-97. 

41 Id. at 401. 

402 Id. Scholars have noted the inconsistency. See, e.g., Alfange, supra note 388, at 400 

(noting tension between Marbury and Cohens); Conrin, supra note 391, at 540-41 (arguing 

that Cohens "abandoned" reasoning of Marbury and left its "precise decision... hanging 

in mid-air"); Currie, supra note 392, at 654-55 (noting that "Marshall himself was to reject 

the implications of his Marbury reasoning in Cohens v. Virginia"). 

403 As with a number of other constitutional provisions discussed earlier, see supra notes 

117-23,131-34,194-209, and accompanying text, it is instructive to note the various wavs in 

which the Court has held that the Supreme Court Jurisdiction Clauses are nonexclusive. 

Cohens stands for the proposition that the Congress may confer on the Court appellate 

jurisdiction over cases falling within its original jurisdiction that may nevertheless have 

originated elsewhere. Thus, although assigned only original jurisdiction over certain cases, 

the Court is not exclusively limited to exercising that form of jurisdiction over them, and, 

despite the arguably contrary language in the Appellate Jurisdiction Clause, the cases described 

as within the Court's appellate jurisdiction are not an exclusive list. In Ames v.


[Vol. 73:1791 

Perhaps then it is appropriate to cede the Marshallian mantle to 

Professor Tribe, since, like the Chief Justice in Marbury, he too seems 

driven by purposes unconnected to a genuine construction of the text. 

Years hence, perhaps he too will reconsider, returning to his own earlier 

view affirming interchangeability. On the other hand, Professor 

Ackerman and I never meant to invoke the Marshall of Marbury in 

favor of the congressional-executive agreement. Because a broad 

construction of the Necessary and Proper Clause plays such a large 

role in the argument, we were invoking the Marshall of McCulloch. 

Now, we might fruitfully add the Marshall of Cohens. At this stage of 

the argument, and in the absence of any compromise solution in the 

offing, 40 4 I can only respectfully submit this dispute over the great 

Chief Justice's approbation to the reader's impartial judgment! 

Kansas, 111 U.S. 449, 467-69 (1884), moreover, the Court extended this ruling to permit 

Congress to assign matters within the Court's original jurisdiction to the lower federal 

courts, and it was already understood that the states had concurrent jurisdiction in some 

cases. Thus, the assignment of original jurisdiction to the Court was not exclusive of other 

courts exercising original jurisdiction in the same cases. 

It is instructive as well to consider where the Court has found and where it has refused 

to find missing "only"s. Thus, while Marbury in effect found a missing "only" in one 

part of the Original Jurisdiction Clause ("Only in all Cases..., the Supreme Court shall 

have original Jurisdiction"), see Marbury, 5 U.S. (1 Cranch) at 174, the Ames Court refused 

to find a missing "only" in another portion of that clause ("In all Cases.... only the 

Supreme Court shall have original jurisdiction over ... "), see Ames, 111 U.S. at 466-69, 

and the Cohens Court refused to find the Marbury "only" in a comparable part of the 

Appellate Jurisdiction Clause ("Only in all the other cases before mentioned, the Supreme 

Court shall have ...") or in still another part of the Original Jurisdiction Clause ("In all 

Cases . .., the Supreme Court shall only have original Jurisdiction ..."), see Cohens, 19 

U.S. (6 Wheat.) at 397-99. I mention the Court's highly contextualized treatment of missing 

"only"s because Professor Tribe makes a great deal of rhetorical hay out of our reliance 

on missing "only's" in the Treaty Clause and Professors Ackerman's and Amar's reliance 

on the same in Article V. See Tribe, Taking Text, supra note 3, at 1244-45 & 1245 0,77 

(ridiculing "the search for absent 'only"'). He seems to think that the interpretive difficulties 

can be surmounted simply by always reading missing "only"s into the text. See id. 

(arguing for default assumption of exclusivity). However, as the Court's original jurisdiction 

jurisprudence demonstrates, this formulaic solution is out of step with the subtleties of 

constitutional interpretation as reflected in our constitutional tradition. Whether a missing 

"only" should be assumed will depend on context and a host of other factors. In any case, 

as should be apparent by now, I have studiously avoided any reliance on missing "only"s in 

the Treaty Clause in making the textual case for the congressional-executive agreement. 

404 Perhaps, we can dub Professor Tribe's view "Marburyan," while Professor Ackerman 

and I retain the label "Marshallian." Admittedly, Professor Tribe's assent may not be 

forthcoming. Should first authorship not count for something 

Imaged with the Permission of N.Y.U. Law Review




Iv 

OF FREE-FORMISM, CONSTITUTIONAL MOMENTS, AND 

RULES OF RECOGNITION 

I hope I have succeeded in demonstrating that Professor Tribe's 

interpretive methodologies cannot withstand critical scrutiny and that 

free-form interpretation is pervasively on display throughout his argument. 

This does not imply that the best construction of the text is 

nonexclusive. As I made clear at the outset, I claim only that the text 

is subject to two conflicting plausible interpretations. For myself, even 

were I to reconsider the persuasiveness of Professor Ackerman's 

larger theory, I would not think it sensible to try to resolve the question 

on purely textual grounds. 4 0 5 The point, however, is that Professor 

Tribe's elaborate arguments, despite the confidence with which he 

presses them and the scorn he heaps on opposing positions, come in 

the end to nothing more than what an astute reader would immediately 

infer from the text. Indeed, Professor Ackerman and I sought to 

exploit this first blush intuition by beginning our article with the question: 

"Whatever happened to the Treaty Clause" 40 6 

But Professor Tribe pushes the free-form theme beyond mere 

textual arguments: He insists as well that loose interpretive methodologies 

penetrate every level of Professor Ackerman's theory of higher 

lawmaking and that our article provides an egregious case-in-point. 4 0 7 

It is worth considering, then, whether his broader theoretical objections 

are more persuasive than the shaky textual points he was able to 

muster. As we shall see, a number of his scattershot arguments are 

simply not meritorious, while others apparently reflect a deep misunderstanding 

of Professor Ackerman's project. Indeed, one of the virtues 

of the latter's efforts at historical reinterpretation is precisely that 

it permits us to avoid the necessity of free-form methodologies in justifying 

the modem legal landscape, methodologies that Professor 

Tribe blithely reaffirms without seeming to notice the deep tensions 

this creates with his supposed interpretive commitments. In still other 

cases, Professor Tribe's arguments provide an occasion to attempt a 

deeper account of the nature of Professor Ackerman's claims. What 

exactly is Professor Ackerman proposing when he urges us to accept 

constitutional change outside the formal requirements of Article V 

45 The methodological questions such an enterprise would raise, however, are beyond 

the scope of this Article. 

406 Ackerman & Golove, supra note 5, at 801. 

407 See, e.g., Tribe, Taking Text, supra note 3, at 1233, 1288 (alleging Professor 

Ackerman takes "a disturbingly loose approach to descriptive and normative matters 

alike" and that our NAFIA article makes manifest "a significant threat to the whole enterprise 

of constitutional dialogue" that is implicit in Professor Ackerman's work).



[Vol. 73:1791 

Does Professor Tribe provide a plausible alternative account Finally, 

Professor Tribe's charges require some defense of our central interpretive 

claim about events in the wake of World War II.40 ° 

Professor Tribe presses a number of points that seem directed at 

some other article and some other theory of constitutional change. 

After attributing various arguments to us, he goes on to give refutations 

which he seems to count as winning debating points on some 

imagined scorecard. However, the arguments he (purports) to refute 

are nowhere to be found within the four corners of our article. Thus, 

he claims that we advocate searching the text for arguable ambiguities 

and view these "as license immediately to leap outside the discourse 

of text and structure; '40 9 that we believe that past violations repeated 

frequently enough justify disregarding contrary textual commands because 

"once the Constitution's text and structure have been ignored 

long enough, we can rationalize away textual commands and the existence 

of a clash between settled practice and long-ignored principle; 

'410 and that we contend that the acquiescence of a branch in the 

violation of structural provisions of the Constitution is sufficient in 

itself to justify the breach despite the fact that it "is not the prerogative 

of those who hold public office ... to abdicate constitutional protections 

essential to the architecture of our government. ' '41x Of 

course, Professor Tribe's arguments, whether they are persuasive or 

not, are beside the point in this context since we did not make any of 

the arguments he chooses to deride. 412 To be sure, Professor 

408 Our claim is that the political events of 1943 through 1948 are best interpreted as a 

self-conscious movement to amend the Treaty Clause informally and validate the new congressional-executive 

agreement procedure. See Ackerman & Golove, supra note 5, at 861- 

96. 

409 Tribe, Taking Text, supra note 3, at 1278; see also id. at 1279 (similar). 

410 Id. at 1285; see also id. at 1280 (similar). For this point, Professor Tribe relies on a 

brief remark Professor Ackerman and I made in a letter to President Clinton regarding the 

WTO Agreement. See id. at 1280 (citing Letter from Bruce A. Ackerman, Professor, Yale 

Law School, and David M. Golove, Professor, University of Arizona College of Law, to 

President William J. Clinton 3 (Sept. 21, 1994) (on file with the New York University Law 

Review)). In context, however, it is clear that the argument was connected to our broader 

argument about the character of the political movement of 1945. In any case, no such 

argument appears in our article. 


412 I believe that the latter two of his arguments are not, in fact, persuasive. Longstanding 

practice is an important data point for constitutional interpretation, and the acquiescence 

of an affected branch, depending upon a number of contextual considerations, may 

in some cases be significant as well. Relevant in the latter case would be, inter alia, the 

degree of consensus among the members of the affected branch, the length of time that 

consensus holds, and whether the purpose of the constitutional provision is to assure political 

accountability precisely in the way the affected branch, by its acquiescence, is seeking 

to avoid. In these respects, among others, I reject Professor Tribe's strict textualism as well 

as other forms of originalism, but I cannot pursue these matters further here.




Ackerman's theory permits constitutional change outside of the formal 

amendment procedures of Article V but only when the strict requirements 

of a specially formulated set of criteria are met. 413 The 

existence of textual ambiguities, past violations, or institutional acquiescence 

are either not among the relevant criteria or are in themselves 

clearly insufficient. Thus, while arguments of the kind Professor Tribe 

pursues make for enjoyable reading, they hardly advance the 

debate. 414 

413 See 1 Ackerman, supra note 4, at 266-90 (elaborating criteria that must be met, in his 

view, to validate an informal amendment). 

414 Professor Tribe also repeatedly chastises us for affirming the constitutionality of 

NAFTA without bothering to analyze its specific terms in any detail. This "reflects precisely 

the free-form character of what Professors Ackerman and Golove put forth as constitutional 

interpretation." Tribe, Taking Text, supra note 3, at 1227; see also id. at 1251-52, 

1277-78 (repeating sharp criticism on this point). Of course, we did not closely analyze 

NAFTA's terms because our argument is that Congress may approve any international 

agreement falling within its substantive powers. Given Congress's power to regulate foreign 

commerce, it is quite evident, and no one has disputed, that NAFTA falls squarely 

within its substantive authority. This should not be surprising. It is frequently unnecessary 

to engage in a searching analysis of the terms of congressional legislation when the only 

question is whether it falls within the scope of one of Congress's enumerated powers. Ironically, 

Professor Tribe's position is closely parallel, but opposite to our own. Despite his 

derisive comments, he, too, eschews analyzing NAFrA's terms. See id. at 1277 (conceding 

as much). Indeed, for him it is unnecessary to determine even whether NAFTA is a regulation 

of foreign commerce. The only relevant fact is that Congress is without power to 

approve international agreements, that is, any international agreement. 

Professor Tribe also argues that our free-form arguments are merely a cover for our 

hostility to the states and that our real goal is to overturn "the Framers' vision of a Union 

of equal and sovereign states." Id. at 1230. I do not know Professor Ackerman's views on 

the continuing values of federalism and admit to some skepticism of my own about the 

value of the Framers' state-centered design in contemporary conditions. However, this is 

obviously entirely beside the point. Professor Ackerman's theory is, and must be, politically 

neutral in the sense that it states criteria for legitimate constitutional transformations 

that apply irrespective of the content of the proposed changes. Depending on events, then, 

we could move either further away from or back towards the Framers' initial vision, just as 

we could move back to an exclusivist view of the Senate's role in treaty-making. The policy 

arguments are irrelevant from this perspective. Professor Tribe displays an ignorance of 

the crucial history, moreover, when he claims that we think it was the "pig-headedness" of 

the states that obstructed the adoption of a formal amendment depriving the Senate of its 

special role. Id. at 1229. On the contrary, all accounts were that the states were more than 

prepared to approve the amendment proposed by the House. The problem was the Senate 

itself. See Ackerman & Golove, supra note 5, at 865 & n.297 (recounting Senate Judiciary 

Committee's determination to delay consideration of treaty amendment). Despite Professor 

Tribe's patriotic sentiments about the Senate's high-mindedness, observers at the time 

were extremely skeptical about whether it could be convinced to cooperate in the elimination 

of its own treasured prerogative. See id. (citing contemporary newspaper editorials). 

When in response to the House's amendment proposal the Senate Judiciary Committee 

cynically announced that it would not consider any amendments until after all the soldiers 

had returned home, the Washington Post declared: "[Tihe Senate will have to be blasted 

out of its foxhole of entrenched power." Signal to the House, Wash. Post, Feb. 28,1945, at 

8.



[Vol. 73:1791 

Professor Tribe also repeatedly suggests that by claiming that Article 

V is nonexclusive, Professor Ackerman's project was inevitably 

doomed to free-formism. Having disregarded the clear meaning of 

Article V, he was bound to feel equally free to find other fundamental 

procedures nonexclusive as well. 415 This is borne out, he thinks, by 

our claims about the Treaty Clause. 416 I leave to Professor Ackerman 

the task of defending the textual argument in favor of a nonexclusive 

reading of Article V.417 For reasons I discuss below, whatever one 

concludes about the success of that endeavor, one cannot defeat his 

larger project simply by rejecting his textual arguments. 418 For present 

purposes, the important point is that Professor Tribe badly mistakes 

what flows inevitably from Professor Ackerman's views about 

informal higher lawmaking. It should be obvious that whether Article 

V is exclusive or not has no bearing on the proper method of engaging 

in textual interpretation of other provisions of the Constitution and 

that nothing in Professor Ackerman's theory seeks to justify or encourage 

loose methods of textual exegesis. What does follow inevitably, 

in contrast, is that there may be times when textual exegesis is not 

the appropriate task for the constitutional interpreter. When an informal 

transformation of a constitutional provision has occurred, the interpreter 

must (at least partially) eschew the text in favor of new legal 

materials that have been validated through non-Article V processes. 

This inevitably requires resort to history but not to free-form interpretive 

methodologies. Hence, our basic thesis about the Treaty Clause: 

Events in 1945 transformed the constitutional status of treatymaking, 

validating a new congressional power to approve international agreements. 

The ultimate justification for this shift-whatever textual arguments 

can be made in its behalf-is based on an interpretation not 

of the original meaning of the text but of the meaning of 1945.419 

415 See, e.g., Tribe, Taking Text, supra note 3, at 1233, 1241 (charging that Professor 

Ackerman's acceptance of an extratextual amendment procedure "has led him to treat all 

constitutional text and structure as casually as he treats Article V"). 

416 See id. at 1233, 1241 (stating that our interpretation of Treaty Clause is 

"[u]nsurprising[ ], and perhaps even predictabl[e]" in light of Professor Ackerman's theory 

of higher lawmaking). 

417 See 2 Ackerman, supra note 4, at 71-88 (analyzing Article V with reference to history 

of Convention). 

418 See infra notes 426-33 and accompanying text. 

419 We noted the indeterminacy of the text only to make it easier for some to accept the 

implications of Professor Ackerman's theory of informal amendment. Our implicit claim 

was that at least when the text is ambiguous, a constitutional movement like that of 1945 

justifies abandoning a long-settled construction of the text for a new reading that is itself a 

plausible construction of the relevant provisions. We likewise thought it might be easier 

for some to accept Professor Ackerman's view in this case because of the peculiar difficulties 

of amending the Treaty Clause under Article V, in which two-thirds of the Senate is 

given a privileged role in validating amendment proposals. See U.S. Const. art. V. Could




There is, however, a deeper flaw in Professor Tribe's argument. 

By insistently focusing on questions of interpretive methodology, Professor 

Tribe misses the main point that Professor Ackerman's is not a 

theory of interpretation at all. His is a theory of exogenous constitutional 

transformation, which, of course, has interpretive aspects and 

consequences, but which leaves untouched many of the fundamental 

methodological questions which are the focal point of Professor 

Tribe's concerns. There is one interpretive consequence of Professor 

Ackerman's view, moreover, that is crucial from the point of view emphasized 

by Professor Tribe, only it works precisely in the opposite 

direction from what he supposes. Rather than encouraging objectionable 

free-formism, one of the central merits of the Ackermanian approach 

is its capacity to liberate us from such methods in justifying 

major parts of the constitutional practices of the modem era. Yet, 

Professor Tribe seems to miss this central point. Even as he reproaches 

us, he unselfconsciously persists in sustaining the constitutional 

transformations of 1937 as unproblematic acts of ordinary 

interpretation that require only cursory examination. 420 This is not 

surprising, for what he emphatically wishes to accomplish is only minor 

constitutional surgery-sacrificing the obscure congressionalexecutive 

agreement without calling into question the more sweeping 

changes of the New Deal. 421 But does anyone doubt that were he so 

inclined, Professor Tribe could mount an impressive original meaning 

assault on the modem Commerce Clause jurisprudence and that the 

textual and structural arguments would be at least as persuasive as 

those he musters against the congressional-executive agreement 

When pressed to explain the difference, he says only that what was at 

stake in 1937 was just "the breadth of Congress's" commerce powers 

two-thirds of that body ever be convinced to renounce through formal amendment the 

prerogative of one-third plus one of its members to veto treaties See Ackerman & 

Golove, supra note 5, at 909 (noting dysfunctionality of Article V in this instance). 

420 See Tribe, Taking Text, supra note 3, at 1296-99 (concluding that New Deal changes 

such as overturning of Lodiner were just "the stuff of ordinary constitutional 

interpretation"). 

421 Professor Tribe does seem to recognize the possibly breathtaking implications of his 

view by suggesting that substantive due process, including not only the modem privacy 

doctrine, but the incorporation doctrine itself, might have to go! See id. at 1297 n.247 for 

one of the truly most remarkable footnotes in the history of legal scholarship. However, 

he comforts himself by thinking that much of what will be lost can be recovered through 

other means, such as reinterpretation of the Privileges and Immunities Clause and expansive 

readings of the prohibition on state bills of attainder. See id. At a minimum, this 

stunning turnabout seems to reflect a major concession to Professor Ackerman's main 

point-our traditional interpretive practices have justified the modem constitution only 

through "free-form" interpretive methods. Notwithstanding Professor Tribe's doubts, 

however, I take no position here on whether modem substantive due process jurisprudence 

can be justified using legitimate, non-Ackermanian interpretive techniques.



which "is a question of degree, not of basic architecture. '422 Yet, this 

seems nothing more than an assertion that so long as no architectural 

provision in Professor Tribe's obscure sense is at stake, the "freeformism" 

that he otherwise condemns is perfectly acceptable. 423 It is 

its ability to avoid the necessity of such interpretive moves that constitutes 

one of the great virtues of Professor Ackerman's theory, and this 

is especially so for formalists like Professor Tribe. Rather than asking 

us to stare at the text so long we can no longer focus on its words or 

the structures it creates, Professor Ackerman directs us to the relevant 

historical moment and urges that we interpret how the people wished 

to change their inherited constitutional practices. 4 2 4 

422 Id. at 1296. Professor Tribe does suggest in passing, but does not defend, the claim 

that the Court's modem jurisprudence is a return to the pre-1880s understanding of the 

Commerce Clause. See id. at 1295. 

423 The same can be said of his explanation of modem substantive due process jurisprudence: 

The overturning of Lochner v. New York, 198 U.S. 45 (1905), see West Coast Hotel 

Co. v. Parrish, 300 U.S. 379, 398-400 (1937), did not require a constitutional amendment 

because it did not represent a rejection of substantive due process. See Tribe, Taking Text, 

supra note 3, at 1296-97 & 1296 n.246. The latter simply took another guise in Griswold v. 

Connecticut, 381 U.S. 479 (1965), where the Court replaced the discredited notion of economic 

liberty with the notion of privacy. See Tribe, Taking Text, supra note 3, at 1296-97 & 

1296 n.246. Leaving aside the timing problems with this argument, it suggests that in Professor 

Tribe's interpretive world, if the Court limits itself to protecting new liberties and 

rejecting old ones, there will never be a need for an amendment. Here again, his striking 

formalism is self-evidently on display: Just as long as we retain the category of substantive 

due process, we are free to use free-form methods as we move from protecting the right to 

contract to the right to an abortion. 

Professor Tribe also applies the same expansive approach to the term "treaty" in the 

Treaty Clause, although he does so only rather softly. See id. at 1247 n.89 (noting "'treaty' 

... might have some evolutionary potential"). To appreciate his point, it is necessary to 

recall his argument that the Treaty Clause only applies to "treaties" and not to lesser 

"agreements," and that the President may conclude unilateral executive agreements only 

insofar as these are not covered by the Treaty Clause. See supra Part III.D.2. His concern 

is that the scope of the President's authority to make unilateral executive agreements (like 

the scope of Congress's powers under the Commerce Clause) has expanded enormously 

from its very modest beginnings in the late eighteenth century and throughout the nineteenth. 

To justify this development, he holds that the meaning of the term "treaty" in the 

Treaty Clause, rather than expanding as time has passed (as in the case of the term "commerce" 

in the Commerce Clause), has instead taken on a narrower and narrower meaning: 

Whereas originally it included virtually all agreements, now it includes only the more important 

among them, since otherwise presidents would regularly be engaged in violations 

of the Treaty Clause. I can perhaps be pardoned for suggesting that this argument is, to be 

charitable, textually disingenuous. 

424 Many have made the point that Professor Ackerman's criteria for valid informal 

amendments may not be definite enough to provide a firm basis for determining when a 

constitutional movement has achieved its goal and that the advantages of textual amendments 

in this respect are of great significance. See, e.g., Michael W. McConnell, The Forgotten 

Constitutional Moment, 11 Const. Commentary 115, 122-40 (1994) (arguing that 

end of Reconstruction would qualify as constitutional moment legitimizing Jim Crow 

under Ackerman's criteria). This is an important objection and one which Professor Tribe 

seems to share. See Tribe, Taking Text, supra note 3, at 1286 & n.216 (endorsing




Professor Tribe also finds free-formism at work in Professor 

Ackerman's "disturbingly loose" conflation of the Founding, Reconstruction, 

and the New Deal, events which Professor Tribe believes 

have little in common. 4 2 s The argument from the illegality of the 

Founding fails, he thinks, on self-reference grounds: Nothing in the 

procedures used for the adoption of the Constitution can tell us 

whether similar procedures are valid under the Constitution itself 

since they simply brought the Constitution, including Article V, into 

force. 426 The argument from the Fourteenth Amendment, in contrast, 

lacks a sense of constitutional proportion. The Civil War was a singular 

event, and even though Article V's requirements were overlooked 

in the adoption of the Fourteenth Amendment (and perhaps the Thirteenth 

as well), the decision to go forward in the face of the opposition 

of the states that had illegitimately made war upon the Union is 

not a precedent for less momentous occasions like the crisis of economic 

regulation of the New Deal or the post-War framing of the new 

international order. 427 These claims constitute a challenge to central 

aspects of the Ackermanian constitutional order and warrant fuller 

treatment than I can give them here. I attempt only the outlines of a 

response, for my pur-pose is less ambitious: to demonstrate that Pro- 

McConnell's analysis). This is yet another way in which Professor Tribe believes that Professor 

Ackerman's theory promotes free-formism. The way Professor Tribe formulates the 

objection, however, is unconvincing. He thinks that only if amendments are strictly limited 

to those that are adopted in accordance with textual procedures can the Constitution 

meaningfully constrain governmental acts; otherwise, "the constraining power of text and 

structure is eroded almost to the vanishing point." Id. at 1280. On the contrary, however, 

there is nothing inherent in an extratextual procedure that undermines the Constitution's 

constraining power. Among other things, its constraining power depends upon the clarity 

of the accepted criteria for extratextual amendments, the scope of any ambiguities in the 

procedure for textual amendments, and the relative determinacy of the textual and nontextual 

amendments that are adopted. If the constitutional tradition in a polity permits extratextual 

amendments in accord with the one-time wishes of any person who can pull 

Excalibur out of a rock, it might have a perfectly determinate constitution, so long as its 

King Arthur rendered only relatively clear constitutional edicts. In contrast, if a polity that 

restricted amendments to those adopted in accordance with a clearly specified textual procedure 

adopted amendments incorporating broadly abstract commands such as "no person 

shall be deprived of liberty or property without due process of the law," the constitutional 

text might be quite indeterminate, and unless it could be made determinate through an 

agreed upon procedure, its constraining power might well be nil. Notwithstanding Professor 

Tribe's way of putting the point, however, it is true that the real question is whether the 

historical and normative resources from which Professor Ackerman draws are adequate to 

provide definite criteria that can meet the general objection. Otherwise, his particular theory 

of extratextual amendment will founder on the shoals of vagueness and may tell us 

only then that nothing succeeds like success. I cannot attempt here to address this larger 

problem. 


426 See id. at 1290-92. 

427 See id. at 1292-94.



[Vol. 73:1791 

fessor Tribe's charge of free-formism is as unwarranted in this case as 

it has been throughout. 

In part, the difficulty in assessing objections like Professor Tribe's 

results from conceptual puzzles at the heart of Professor Ackerman's 

theory; in part, these ambiguities are the consequence of the style of 

his argument which presses different conceptual approaches simultaneously. 

But it will be helpful to sort out at least two fundamental 

questions: the level at which the theory purports to operate and the 

relationship between its descriptive and normative claims. To put the 

matter somewhat differently, to what extent does it rely upon interpretive 

materials that are endogenous or are exogenous to the Constitution 

in arguing for the nonexclusivity of the amendment procedure 

And to what degree does the theory rest on the descriptive claim that 

it provides the most accurate account of our actual constitutional 

practices rather than the view that it provides the most attractive normative 

account while still fitting our actual practices 

Begin with two views of Professor Ackerman's claims that seem 

to offer the most plausible constructions of his project. The first takes 

the endogenous claims as the core of the argument. Under this view, 

the theory rests upon the thesis that the Constitution itself is best interpreted 

as providing for amendment procedures that are not explicitly 

identified in the text. This is the case, moreover, solely on the 

basis of interpretive materials that are internal to the Constitution according 

to some plausible theory of interpretation. In this version, the 

normative claims appear to be at least as important as the descriptive. 

They provide a reason why we should accept practice arguably in conflict 

with the text as controlling: because, in Dworkin's sense, that 

practice reflects the most attractive account of the Constitution and 

still fits our history. 428 

The second construction of Ackerman's theory views it as resting 

principally on an exogenous argument about the character of the ultimate 

rule of recognition in our society. The claim is that irrespective 

of the text, the rule of recognition establishes that our fundamental 

law-which we refer to broadly as the Constitution-can be altered 

outside Article V when the public adequately manifests its considered 

will. Rules of recognition are purely matters of social and political 

fact-the rule of recognition is just whatever the relevant officials ac- 

428 See Dworkin, Freedom's Law, supra note 2, at 11 (arguing judges should seek morally 

best conception of constitutional principles that still fits historical record); Dworkin, 

Law's Empire, supra note 2, at 62 (arguing that we should seek interpretation that makes 

artistic work or social practice or structure "the best it can be" consistent with constraints 

of history).




cept as the source of validation for all binding law. 429 So in this version 

of the argument, the descriptive claim takes on far greater 

significance. The claim is that the relevant officials in this country 

have accepted in the past, and continue to accept, something like Professor 

Ackerman's criteria as the basis for alterations of the fundamental 

law. This does not entail that the normative features of the 

argument are irrelevant. But from this vantage point, Professor 

Ackerman's writings appear to be an effort to make the terms of our 

rule of recognition more perspicuous to ourselves and to urge only 

their refinement on political/normative grounds. 430 Thus, while it is 

critical that he not simply be making a normative plea for a new rule 

of recognition, nothing precludes him from urging a honing of the 

marginal features of our existing rule to make it more consistent with 

normative political theory. 

To assess Professor Tribe's objections, we need to evaluate them 

in light of these two constructions of Professor Ackerman's project. 

Begin with the endogenous interpretation, which seems to be the view 

Professor Tribe has principally in mind. 431 Is there a plausible account 

429 See H.L.A. Hart, The Concept of Law 97-114 (1961) (articulating his famous positivist 

theory of law); see also Frederick Schauer, Amending the Presuppositions of a Constitution, 

in Responding to Imperfection, supra note 17, at 145, 149-52 (endorsing Hart's view). 

430 Perhaps, too, his project is an effort to explain why those who may wish to undo 

important parts of the New Deal and post-War constitutional orders cannot base their 

arguments on fidelity to some long-accepted tradition. They are involved in a political 

effort either to amend the fundamental law or to alter our rule of recognition itself. Textual 

fidelity is not a politically neutral interpretive methodology inherent in our constitutional 

tradition. 

431 See Tribe, Taking Text, supra note 3, at 1291-92 & 1291 n.227 (taking exception to 

claim "that the text of the Constitution is rightly seen as itself allowing alternative modes of 

amendment... as a matter of constitutional law"); see also Schauer, supra note 429, at 147 

(reading Professor Ackerman as making only an endogenous argument for nonexclusivity 

of amendment process). Professor Tribe's narrow objections to Professor Ackerman's internal 

arguments seem implicitly to endorse the validity of his exogenous claims. But this 

would seem to undermine seriously Tribe's position. Whatever conceptual differences 

there may be between the endogenous and the exogenous views, they do not matter much 

from the perspective underlying Professor Tribe's objections. Thaus, even were Professor 

Tribe right that Professor Ackerman can rely only on the exogenous argument, the implications 

of the latter's views would be the same: The Constitution can be and has been 

amended outside the formal requirements of Article V. It makes no difference in the final 

analysis whether these "amendments" are technically valid "as a matter of constitutional 

law" or only are treated as such by the relevant public officials in accordance with our 

ultimate rule of recognition. Either way, the Constitution has been informally amended 

and those amendments are as binding on the courts and the political branches as if they 

had been adopted in accordance with Article V. It is particularly odd, then, that Professor 

Tribe self-consciously aligns himself with Professor Schauer in this respect. See Tibe, Taking 

Text, supra note 3, at 1291 n.227 (endorsing Schauer). Professor Schauer is quite explicit 

in holding that the Constitution can be amended outside the requirements of Article 

V if our rule of recognition allows it to be, see Schauer, supra note 429, at 156-57, and that 

as a descriptive matter this is almost certainly the case, see id. at 156-58 (noting that exclu-



[Vol. 73:1791 

of how the "illegality" of the Constitution's adoption under the Articles 

of Confederation bears on the proper internal interpretation of 

Article V Professor Tribe is of course right that the procedure used 

to validate the Constitution cannot conclusively demonstrate the 

proper interpretation of Article V. The Constitution was not, and 

could not, have been adopted pursuant to Article V which was not 

part of the Articles and did not itself become valid law until the Constitution 

went into force. But the fact that the Founding cannot be 

relevant in this respect does not entail that it provides no support for 

the endogenous interpretation in other respects. On the contrary, the 

argument is that the Framers' self-conscious end-run around the 

amendment provision of the Articles reflected their principled view 

that formal textual requirements cannot and should not be controlling 

in the face of deep and broad manifestations of popular will, a view 

apparently affirmed by the American people in their willingness to 

accept the Constitution's validity despite the evident violation of the 

Articles. Given this view, there is good reason to suppose that in 

adopting Article V the Framers were not attempting to contain future 

expressions of popular sovereignty through precisely the kind of formalist 

means they had just eschewed. 

The conceptual issues under the endogenous view become somewhat 

stickier when we turn to the Fourteenth Amendment. On its 

face, Professor Tribe's principal argument seems to be that the constitutional 

disruption caused by the Civil War is simply on a different 

order of magnitude than the New Deal crisis or the conflicts of the 

second World War. Professor Tribe apparently believes that it was 

justified to leap out of the text only in the first, not in the latter cases. 

However, in the first place, it is unclear by what metric one would 

compare the varying levels of crises at these momentous historical 

junctures. And, in any case, since deep and abiding shifts in constitutional 

law occurred in each, it is unclear why the differences should be 

of significance in this context-unless Professor Tribe means to accept 

Professor Ackerman's claim validating the concept of informal 

amendments and wishes only to state a different normative judgment 

about where the line should be drawn. 

sive amendment procedure is "almost certainly not the American approach" and stating 

that descriptively his view "is that the American legal and constitutional culture treats the 

procedures of Article V as presumptively but not conclusively constraining"); see also Kent 

Greenawalt, The Rule of Recognition and the Constitution, 85 Mich. L. Rev. 621 (1987) 

(expressing view similar to Professor Schauer's). If the congressional-executive agreement 

is valid in accordance with our rule of recognition, then precisely what is Professor Tribe 

saying when he brands it unconstitutional Is it unconstitutional but valid and binding on 

the courts and political branches If so, it would certainly have been helpful had he said 

SO.




Perhaps we can reformulate Professor Tribe's argument to bring 

out what may be his deeper objection, that the Fourteenth Amendment 

is somehow categorically different from the New Deal and 

World War II. The Fourteenth Amendment is valid "law," he may 

think, only because the trauma of the nation's sundering led to a 

change in our ultimate rule of recognition: Our post-Civil War rule of 

recognition, he might claim, directs officials to apply the Fourteenth 

Amendment as such, along with the provisions of the Constitution, as 

the fundamental law of the land. In contrast, the transformations of 

the New Deal and World War II were purely matters of interpretation 

under the existing Constitution and must stand or fall on the merits of 

what they purported to be-internal interpretations of the text. There 

are two replies to this version of the objection. First, Professor Tribe's 

point can be made in reverse. The Reconstruction Congress did not 

purport to change the rule of recognition, just to amend the existing 

Constitution. It chose, after all, to denominate the Fourteenth 

Amendment an "amendment." This choice suggests that the Reconstruction 

Congress did not believe Article V constituted the sole 

means of changing the Constitution and that there are times when the 

will of the nation as a whole outweighs the conflicting interests of the 

states as such. On this reading, the adoption of the Fourteenth 

Amendment provides substantial support for Professor Ackerman's 

claim that we have not as a nation understood Article V to state the 

exclusive internal means of amending the Constitution. Second, the 

New Deal and World War II effected tectonic shifts in our constitutional 

practices that have proved long-lasting and durable. Even 

though the changes wrought were based on at least highly arguable 

textual premises, our officials have shown virtually no serious interest 

in revisiting the fundamental questions resolved in those historical periods. 

Indeed, this deference has persisted even with the rise of radically 

new interpretive methodologies under which the earlier 

decisions are highly vulnerable. 432 Professor Tribe himself witnessed 

the resilience of these practices when the Senate cursorily disregarded 

his strenuous constitutional objections to the VTO Agreement, despite 

the fact that he was only pressing that body to defend its own 

jealously guarded prerogative of yesteryear. 433 So, while it is true that 

432 In this respect, United States v. Lopez, 514 U.S. 549 (1995), seems like a rather weak 

shift in direction, not a promise of a wholesale repudiation of the New Deal Commerce 

Clause jurisprudence. See id. (striking down statute that prohibited carrying firearms in 

areas near schools as beyond Congress's powers under Commerce Clause). 

433 See supra notes 42-44 and accompanying text. The Senate approved the VTO 

Agreement as a congressional-executive agreement despite Professor Tribe's ominous 

warning that doing so would effectively make it impossible for it to reassert its authority in



[Vol. 73:1791 

no formal amendments were adopted, our political practices suggest 

that we have given these highly problematic "interpretive" changes a 

similar stature and that they are, in this respect, far more comparable 

to the Fourteenth Amendment than Professor Tribe is willing to 

admit. 

The relevance of the Founding and the Civil War to the exogenous 

view is, I think, even more readily apparent. The fact that the 

Framers did not feel bound to respect the textual requirements of the 

Articles when they could obtain widespread popular support for their 

new constitutional order suggests the deep roots of antiformalism in 

our political culture. The acceptance of the Constitution as valid and 

binding confirmed that our rule of recognition directs officials sometimes 

to validate as fundamental law provisions not adopted in accordance 

with the governing text. Furthermore, continuity with the 

Founding period in this respect was demonstrated during Reconstruction 

when the Fourteenth Amendment was immediately accepted as 

valid despite the evident inconsistencies with Article V. The tradition 

was carried forward yet again when officials accepted the core New 

Deal constitutional order and its rippling implications immediately after 

the 1937 switch. All of these instances, then, are evidence that our 

rule of recognition directs officials to accept some "amendments" to 

the fundamental law that are not adopted in accordance with textual 

strictures. It also suggests that these "amendments" are to be validated 

when a political crisis has mobilized large portions of the public 

decisively in favor of changes to the existing constitutional order. 

Certainly, this is not the only plausible construction of these historical 

events, but as I indicated at the outset, my purpose has not been to 

give a full defense to Professor Ackerman's views but only to acquit 

them of Professor Tribe's charge of free-formism. 

Finally, Professor Tribe finds free-form methods at work in our 

interpretation of the events during the mid-1940s that are at the core 

of our argument. In his view, there was no "constitutional moment"; 

by acceding to the congressional-executive agreement, the Senate just 

made a strategic maneuver to avoid a constitutional amendment stripping 

it entirely of its unique treaty role. 434 The elites in Washington, 

moreover, were fully cognizant that they were violating the Constitution. 

That is why the House adopted an amendment proposal. In this 

respect, he thinks, 1945 was entirely different from events in 1787, the 

the future. See GATr Hearings, supra note 41, at 301-02, 311 (testimony of Laurence H. 

Tribe, Professor, Harvard Law School) (admonishing Senate that adoption of GATr 

"outside the strictures of the Treaty Clause could spell doom for the continued viability of 

the Treaty Clause itself"). 

434 See Tribe, Taking Text, supra note 3, at 1284-85.




1860s, and 1937. 435 In any case, Professor Ackerman and I have gotten 

the constitutional moment all wrong; all the people did in 1945 

was engage in a kind of extratextual popular approval of the United 

Nations Charter. Their principal concern was not with the Treaty 

Clause at all. 436 

Here, Professor Tribe is finally arguing on our territory, but his 

arguments are as unpersuasive in this context as elsewhere. First, it is 

unclear what to make of his claim that the collapse of the Senate's 

resistance was a strategic move. Of course it was; that is precisely the 

point of our argument. But it was not just a strategic move; it was 

taken with full cognizance of the depth and breadth of public opinion 

arrayed against the Senate's traditional position-the decisive shift towards 

internationalism and the widespread association of a discredited 

isolationism with the Senate and its historical treaty role-all as 

reflected in the outcome of the elections of 1944V37 Furthermore, 

Professor Tribe seems not to appreciate the disturbing implications of 

his view. For he seems to be suggesting that there are no consequences 

to the Senate's deliberate derailment of the "regular" amendment 

processes through temporary acquiescence in an 

unconstitutional procedure. A thought experiment (not too far from 

historical reality) will bring out the objection: Imagine the position of 

the conservative members of the Supreme Court in 1937 under the 

following scenario. The country, they sense, is close to opting for formal 

amendments that would sweep aside the principles of limited national 

government and laissez-faire economics to which these justices 

are deeply committed. To avoid this unappealing result, they decide 

to "switch" in the short-run, yielding to New Deal legislation that in 

their view is unconstitutional. But their acquiescence is only strategic; 

they simply abide the right moment-when the economic crisis has 

lessened and the people are no longer mobilized. Then they will 

"switch" again, this time to reinstate the principles that they had apparently 

forsaken in their earlier decisions. Is Professor Tribe such a 

formalist that he would accept this constitutional fraud Might even 

he see a way to uphold a practice that apparently would have made it 

into formal constitutional law but for the cynical course adopted by 

the Court Yet, this is precisely what the Senate, in his view, did in 

1945.438 More generally, when Franklin Roosevelt made the decision 

to forego the formal amendment track in 1944-45 and opt for informal 

435 See id. at 1285, 1301. 

436 See id. at 1284 n.210. 

437 See Ackerman & Golove, supra note 5, at 861-96 (detailing pertinent history). 

438 For assessments of the popular support for an amendment, see id. at S62-64. For an 

account of the Senate's retreat, see id. at 889-96.



[Vol. 73:1791 

constitutional transformation, he was relying upon a particular jurisprudential 

background-the then-ascendant view that accepted common 

law methods in constitutional law, at least when combined with 

deep shifts in popular sentiment. After United States v. Darby, 439 

Wickard v. Filburn, 440 and United States v. Pink, 441 to name but a few 

cases, Roosevelt could easily have concluded-indeed, he would have 

been a fool to think otherwise-that the Supreme Court would agree 

that the congressional-executive agreement was "constitutional." 

Why then force an amendment through the Senate, jeopardize bipartisan 

internationalism, and distract the country from the monumental 

substantive problems of the new post-War order, all to satisfy a formalist's 

sense of propriety It is in this context that the deeply reactionary 

implications of Professor Tribe's methodologies become 

evident. 

Second, Professor Tribe thinks the fact that the political leaders in 

Washington were aware that the congressional-executive agreement 

was inconsistent with the traditional understanding of the Treaty 

Clause somehow delegitimates their constitutional movement and 

sharply distinguishes it from the Founding, the Civil War Amendments, 

and the New Deal transformations of 1937. Professor Tribe, 

however, seems to have the point precisely in reverse. We have something 

to worry about when our elites are unaware that the movements 

they lead are actually for constitutional transformation. For then we 

do not have deliberate movements for constitutional change but political 

movements founded on mistaken premises. We would have good 

reason to fear that whatever changes were proposed had not received 

the kind of deliberative consideration that is an essential predicate for 

accepting a shift in our constitutional practices. 442 It is obscure, moreover, 

why Professor Tribe believes that this awareness distinguishes 

the earlier crises. On the contrary, the historical evidence is quite 

clear that the Framers were well aware that they were disregarding the 

439 312 U.S. 100 (1941) (upholding Fair Labor Standards Act as valid exercise of commerce 

power). 

440 317 U.S. 111 (1942) (upholding Agricultural Adjustment Act as valid exercise of 

commerce power). 

441 315 U.S. 203 (1942) (expanding scope of President's independent foreign affairs authority 

by upholding Litvinov Assignment). For a discussion of the impact of Pink and 

other decisions on the President's foreign affairs powers, see supra note 376. 

442 See 1 Ackerman, supra note 4, at 285-88 (arguing that crucial requirement for validating 

informal amendment is deliberative character of support for change in constitutional 

practice). In fact, if there is a concern about 1945 it should be the degree to which 

the public might have been misled about the nature of the change they were being asked to 

endorse.


December 1998] AGAINST FREE-FORMS FORAIALISM 

1939 

requirements of the Articles of Confederation, 4 3 that the Reconstruction 

Congress had no illusions (and little concern) about inconsistencies 

with the formal requirements of Article V, 4 " and that the New 

Dealers were as aware of the constitutional infirmities of their efforts 

as the later New Internationalists were of their own." s 

Third, Professor Tribe's interpretive claim that 1945 was really 

about popular ratification of the United Nations Charter is no more 

persuasive. He simply offers this as a contrasting hypothesis but provides 

no historical evidence to support his view. His claim is nevertheless 

important in another sense. I agree that 1945 is best read as an 

affirmation both of the congressional-executive agreement and of the 

principles enshrined in the Charter (and other foundational elements 

of the post-War era). The latter point, however, rather than having 

the narrow import Professor Tribe implicitly attributes to it, is in fact 

rich with constitutional significance. Although I cannot defend the 

point here, much of the legitimacy of the post-War constitutional order 

in foreign affairs rests upon changes that were achieved by the 

sweeping political movement behind the Charter. Not least in this respect 

are its implications for the modem division of authority over the 

war powers. 446 

CONCLUSION 

I have been arguing in part against free-form formalism, but this 

is admittedly a rather easy target. My real efforts have had a deeper 

purpose-to illustrate the weaknesses of Professor Tribe's textualist 

interpretive methodology and the tendency of strict formalist approaches 

to self-defeating indeterminacy. From Professor Tribe's 

steadfast effort to draw the textual line against undisciplined interpreters, 

I have argued, comes fresh reason to doubt any methodology 

that focuses single-mindedly on the text. 

Of course, constitutional theory has not been my only, or even 

necessarily my most pressing concern. I have also felt an urgent 

need-and responsibility-to reply to Professor Tribe's all-out textual 

assault on the congressional-executive agreement. Despite his protestations 

to the contrary, the modern practice has become as essential to 

the conduct of our foreign affairs as the Federal Reserve is to the reg- 

443 See I id. at 167-79 (reviewing Federalist Papers on point); 2 id. at 49-65 (revieing 

history of Convention and contemporaneous events). 

444 See 2 id. at 99-252 (reviewing history of Reconstruction). 

445 See 2 id. at 255-382 (reviewing history of New Deal). 

446 For an account of the constitutional controversy over the Charter and the wvar powers, 

see Jane E. Stromseth, Rethinking War Powers: Congress, the President, and the 

United Nations, 81 Geo. L. 597 (1993).



ulation of our banking and monetary systems. I have therefore sought 

to defend the claim that the text can plausibly and persuasively support 

either the original or the modem understanding. But this is necessarily 

only a limited defense of the congressional-executive 

agreement. Although compatibility with the text is important in virtually 

anyone's book, the text's indeterminacy is not sufficient on its 

own to ground the constitutionality of the modern procedure. Something 

more will be needed to convince those who find themselves unable 

to affirm the kind of constitutional theory that Professor 

Ackerman and I have proposed. So, I have contented myself here 

with a more modest goal-opening up at least some interpretive space 

from which defenders of the congressional-executive agreement can 

launch a more conventional constitutional counter-attack. 

In our article, Professor Ackerman and I sought to pose a challenging 

dilemma to the community of constitutional lawyers and 

scholars: A basic and long-accepted constitutional practice in the field 

of foreign affairs turns out, on close inspection, to be extremely difficult 

to justify under the ordinary forms of constitutional argument. 

Contrary to common conceptions, its roots, we claimed, are found not 

in a long accretion of slowly expanding precedent, but in a revolutionary 

and self-conscious moment of constitutional transformation. Assuming 

that the practice itself is a fixed point in our shared 

constitutional universe, we invited others to consider whether this history 

suggested the need for rethinking some of the received conventions 

of our field. 44 7 Though provoked by our argument, Professor 

Tribe ultimately declined our invitation and instead sidestepped the 

dilemma we posed. Donning the mask of the constitutional purist and 

invoking a rhetoric charged with patriotic overtones, he simply denied 

that there is a dilemma-if the modem practice cannot be justified 

under conventional interpretive methods, then it is simply and ipso 

facto unconstitutional, the consequences notwithstanding. Worse, he 

berated us for having failed to pose the dilemma starkly enough! Despite 

our claim that the original intent and the first hundred and fifty 

years of practice were unequivocal, our concession that the text is indeterminate 

was blatantly wrong and revealed our true status as freeformists 

bent on undermining the rigorous legal science underlying 

the discipline of constitutional law. 

In pragmatic human affairs, purism is rarely helpful or compelling, 

and Professor Tribe's above-the-fray approach to the 

447 This is a kind of "reflective equilibrium" argument. See John Rawls, A Theory of 

Justice 19-21 (1971) (articulating role of reflective equilibrium in normative politicaleconomic 

theory).




congressional-executive agreement is no exception. The imperatives 

of history and contemporary realities play an undeniable role in our 

constitutional debates, and no amount of purist posturing or affectation 

of superior moral rectitude can alter these basic facts of constitutional 

life. We will always have to struggle over the balance between 

pure principle and changing times; neither can achieve total dominance. 

Nor is constitutional law mathematics or the text a place to 

which we can retreat to escape the value conflicts that inevitably characterize 

the views of citizens in a pluralistic society. Constitutional 

interpretation is messier than that, more inclusive, less certain, and 

almost always legitimately open to struggle. 

Professor Tribe is surely cognizant of these platitudes-indeed, 

he has made them himself on past occasions. 448 Perhaps then, his sudden 

turnabout reflects a perceived need to shore-up the solidity of 

both the New Deal Revolution and the Warren Court civil liberties 

jurisprudence. He may believe that by drawing the line at the (in his 

view) less centrally important congressional-executive agreement he 

can mollify the critics. If this is his underlying aim, however, it misses 

the mark. There is no denying the parallels between the 

congressional-executive agreement and, for example, the modem 

Commerce Clause jurisprudence. By calling into question the validity 

of the former, however, he will only have fueled the fires of those who 

have led the charge against the New Deal Revolution and the Warren 

Court precedents. 449 With all due respect, I suspect that few will be 

impressed by his efforts to support the textual case for the modem 

Commerce Clause while denigrating the textual argument for the congressional-executive 

agreement. 450 If promoting reconciliation underlies 

his view, then he has, in my view, missed the main reconciliatory 

virtue in the congressional-executive agreement, which is just a cautionary 

point: The modem post-1937 constitutional landscape has 

something of value for everyone, and it may prove impossible to aban- 

448 See Tribe, Choices, supra note 39, at 3-6: 

Even if we could settle on firm constitutional postulates, we would remain 

inescapably subjective in the application of those postulates to particular 

problems and issues.... 

Anyone who insists, for instance, that "fidelity to text" must be the core 

commitment of a constitutionalist must confront the indeterminacy of text and 

must justify giving to one or another vision of language such binding force over 

our lives. 

449 See, e.g., United States v. Lopez, 514 U.S. 549, 584, 601 (1995) (Thomas, J., concurring) 

(stating that Court should "temper" its Commerce Clause jurisprudence in future and 

return to "the original understanding of that Clause"). 




[Vol. 73:1791 

don those parts we revile, or to which we are indifferent, without 

throwing into doubt those we cherish. 

I began this Article with a tribute to Professor Tribe's monumental 

achievements as a constitutional scholar, and nothing I have said is 

meant to cast doubt on his preeminent stature. But his latest work 

does present a puzzle: Why suddenly posture himself as a hard-nosed, 

rigorous textualist, unmoved by controversial value predilections and 

political ideals Is it-and here I freely speculate-the persistent 

voice of his critics finally provoking him to show somehow that they 

have it all wrong That would, in my view, be a mistake. They are not 

literally wrong in charging him, as they have, with the creative use of 

constitutional doctrine. Surely, he has been an enormously creative 

scholar-but creative in the best sense of the term. His creativity has 

produced an immensely valuable body of work that has been, and no 

doubt will continue to be, tremendously influential in shaping our constitutional 

law. So, there is no reason to accept their critiques or to 

bend his own methods to their demands-no reason to jump, even if 

only half-heartedly and only to a limited extent, on the illusory 

originalist bandwagon. Such a move, as I have sought to show, can 

only undermine, not advance, the legitimacy of his constitutional 

project. 

Professor Tribe has thus failed to advance the discussion of the 

congressional-executive agreement past the point where Professor 

Ackerman and I left it in our earlier article, and that is where it remains 

today. Perhaps the debate will pick up again some time in the 

future. One thing, however, is certain. The rich historical and interpretive 

story of the Treaty Clause will continue to provide constitutional 

theorists with a remarkable range of materials for considering 

the controversial questions that continue to vex the field.

Against Free-Form Formalism - NYU Law Review

You also want an ePaper? Increase the reach of your titles

Delete template?

Save as template?